test(video): reproduce Kiro malformed payload retention

changelog.d/fixes/11658-video-transcript-log-redaction.md

@@ -0,0 +1 @@
+- **fix(security):** Video Bridge transcript text is now omitted from OmniRoute-owned retained request, response, usage, error, stream, continuation, handoff, and Memory surfaces while the live bridge request remains intact. Server-generated descriptions are matched by bounded SHA-256/length identities emitted only by a successful Video Bridge rewrite; ordinary fields named `transcript` and caller-forged description prose remain untouched. Traversal is cycle-safe and bounded, and hostile getters, proxies, or budget overflow fail closed to a constant omission marker instead of leaking content or breaking the request. Custom plugins and guardrails remain privileged processors of the live payload and must secure any sinks they create themselves. ([#11658](https://github.com/diegosouzapw/OmniRoute/issues/11658))

docs/security/GUARDRAILS.md

@@ -1,13 +1,13 @@
 ---
 title: "Guardrails"
-version: 3.8.50
+version: 3.8.51
-lastUpdated: 2026-08-24
+lastUpdated: 2026-08-27
 ---
 
 # Guardrails
 
 > **Source of truth:** `src/lib/guardrails/`
-> **Last updated:** 2026-08-24 — v3.8.50 (Video Bridge visual dedup hardening + focused captions)
+> **Last updated:** 2026-08-27 — v3.8.51 (Video Bridge transcript-retention boundary)
 
 Guardrails enforce safety, policy, and content transformations at the boundary
 between OmniRoute and upstream providers. Each guardrail can inspect (and
@@ -439,6 +439,69 @@ are rendered as untrusted observations alongside the frame captions. Invalid,
 out-of-range, or provenance-free text is rejected rather than mixed into the
 caption stream.
 
+#### Transcript retention boundary
+
+Transcript text remains available to the live bridge and upstream request, but it is not retained
+by OmniRoute's logging and memory surfaces. `videoTranscriptLogRedaction.ts` recognizes
+`transcript` and `audioTranscript` only inside supported video carriers. A field with either name
+in an ordinary tool argument or application object is preserved, and delimiter-shaped caller
+prose is not trusted as a bridge result.
+
+When the Video Bridge produces a description containing validated cues, it emits only a
+SHA-256/length identity in guardrail metadata. The request pipeline accepts those identities only
+from a successful, payload-modifying `video-bridge` result and uses them to omit the exact generated
+description segment after translation or concatenation. Raw transcript text is never placed in the
+metadata.
+
+The retention policy covers raw, converted, and provider request logs; provider and client response
+bodies; detailed pipeline artifacts; call logs, active/pending usage, rejected requests, proxy/error
+diagnostics, SSE chunks, and Memory extraction. Response bodies and stream chunks for a sensitive
+request are omitted because an upstream may echo request text. Safe operational metadata such as
+status, timing, provider, model, headers-presence, token counts, and the fact that redaction occurred
+remains available. The constant retained marker is `[omitted: video transcript]`.
+
+Sensitivity is resolved before the guardrail chain from the original client body and resolved again
+after the chain from the original body, processed body, and trusted description identities. A
+guardrail that removes or replaces the raw carrier therefore cannot clear the request-scoped bit.
+Explicit video parts remain sensitive when malformed, and nested carrier objects are inspected under
+the same bounds. Rejected-request persistence also inspects the body itself, so forgetting to pass the
+bit at one caller does not expose a raw carrier.
+
+The request-scoped bit protects OmniRoute-owned application paths in addition to structured payload
+clones. Native priority, round-robin, pinned, and fusion diagnostics; quality-rejection and live-event
+logs; terminal combo errors; proxy fast-fail logs; guardrail-registry logs; plugin-dispatcher failure
+logs; and stream lifecycle callbacks retain only the constant marker when their detail could echo the
+request. A request-scoped executor logger keeps the diagnostic tag but replaces arbitrary executor
+messages with the marker and drops attached metadata before those values reach the application
+logger. Context Relay and Universal Handoff summaries are still generated from a sanitized clone:
+safe adjacent conversation context remains available while recognized carriers and trusted generated
+descriptions are omitted. Memory may likewise extract safe request-derived text from the sanitized
+copy, but response-derived Memory extraction is skipped because an upstream response may echo the
+transcript. Daily-quota classification still inspects the original upstream error, while sensitive
+requests use the server-resolved model as the retained lockout key instead of promoting a model token
+parsed from that diagnostic. These rules affect retained copies, not the live provider request or the
+functional response returned to the same client.
+
+Inspection is bounded by depth, aggregate entries, trusted identities, description-prefix
+occurrences, candidate hashes, and total hashed text. Cycles and binary views are handled without
+retaining their contents. If a bound is exceeded, or an enumerable getter/proxy throws during
+inspection, the retained copy fails closed to the omission marker; the live request is not changed.
+
+Privacy-redacted Responses artifacts cannot reconstruct the omitted transcript portion of
+`previous_response_id` history. `responsesContinuationStore.ts` therefore reuses only valid,
+non-truncated retained `input` and `output` arrays, including the explicit omission marker, and never
+attempts to restore raw transcript text from the pipeline marker. This preserves continuation for the
+safe portions of the turn while making the missing private segment visible; a client that needs that
+context again must submit it as part of a new live request.
+
+Custom guardrails and plugins are privileged, in-process processors: they intentionally receive the
+live payload so they can inspect, transform, or block it. OmniRoute supplies the server-owned
+`videoTranscriptSensitive` bit, wraps `GuardrailContext.log`, and protects errors emitted by its native
+plugin dispatcher. It cannot control a third-party processor that independently writes the live body
+to `console`, a file, a database, or the network. Operators must audit such code and treat its own
+sinks as outside the OmniRoute-owned retention boundary; clearing or ignoring the bit does not make
+the transcript safe to persist.
+
 An advanced caller may provide an already-authorized `audioTranscript` track
 for the same video. The fusion seam runs visual and audio observations under
 one deadline and abort signal, orders them on a common timeline, collapses

open-sse/executors/base.ts

@@ -179,10 +179,10 @@ export type ProviderCredentials = {
 };
 
 export type ExecutorLog = {
-  debug?: (tag: string, message: string) => void;
+  debug?: (tag: string, message: string, data?: Record<string, unknown> | null) => void;
-  info?: (tag: string, message: string) => void;
+  info?: (tag: string, message: string, data?: Record<string, unknown> | null) => void;
-  warn?: (tag: string, message: string) => void;
+  warn?: (tag: string, message: string, data?: Record<string, unknown> | null) => void;
-  error?: (tag: string, message: string) => void;
+  error?: (tag: string, message: string, data?: Record<string, unknown> | null) => void;
 };
 
 export type ExecuteInput = {
@@ -202,6 +202,10 @@ export type ExecuteInput = {
    * this to apply client-format-aware policies such as `</think>` close-marker
    * suppression. */
   clientResponseFormat?: string | null;
+  /** True when upstream diagnostics may echo a video transcript. Executors must
+   * preserve operational responses/errors while omitting those echoes from
+   * retained logs. */
+  videoTranscriptSensitive?: boolean;
   /** Callback to persist tokens that are proactively refreshed during execution.
    * Accepts a partial credentials patch (e.g. `{ accessToken, refreshToken }` or
    * `{ testStatus: "expired", isActive: false }`); the caller merges into the

open-sse/executors/glm.ts

@@ -1,4 +1,5 @@
 import { randomUUID } from "node:crypto";
+import { redactVideoTranscriptSensitiveText } from "../../src/lib/guardrails/videoTranscriptLogRedaction.ts";
 import type { KeyHealth } from "../services/apiKeyRotator.ts";
 
 import { DefaultExecutor } from "./default.ts";
@@ -45,6 +46,11 @@ function asRecord(value: unknown): JsonRecord | null {
   return value && typeof value === "object" && !Array.isArray(value) ? (value as JsonRecord) : null;
 }
 
+function retainGlmDiagnostic(error: unknown, input: ExecuteInput): string {
+  const message = error instanceof Error ? error.message : String(error);
+  return redactVideoTranscriptSensitiveText(message, input.videoTranscriptSensitive === true);
+}
+
 function getEffectiveKey(credentials: ProviderCredentials): string {
   const extraKeys = (credentials.providerSpecificData?.extraApiKeys as string[] | undefined) ?? [];
   if (credentials.apiKey && credentials.connectionId && extraKeys.length > 0) {
@@ -465,6 +471,7 @@ export class GlmExecutor extends DefaultExecutor {
         timeoutMs: STREAM_READINESS_TIMEOUT_MS,
         provider: this.provider,
         model: input.model,
+        redactUpstreamDiagnosticForLog: input.videoTranscriptSensitive === true,
         log: input.log,
       });
       response = readiness.response;
@@ -559,7 +566,7 @@ export class GlmExecutor extends DefaultExecutor {
       if (!isRetryableGlmFallbackError(error)) throw error;
       input.log?.debug?.(
         "GLM_FALLBACK",
-        `${primaryTransport} error (${error instanceof Error ? error.message : String(error)}); trying ${fallbackTransport}`
+        `${primaryTransport} error (${retainGlmDiagnostic(error, input)}); trying ${fallbackTransport}`
       );
     }
 
@@ -572,7 +579,7 @@ export class GlmExecutor extends DefaultExecutor {
       if (!primaryResult) throw error;
       input.log?.debug?.(
         "GLM_FALLBACK",
-        `${fallbackTransport} fallback failed (${error instanceof Error ? error.message : String(error)}); returning primary response`
+        `${fallbackTransport} fallback failed (${retainGlmDiagnostic(error, input)}); returning primary response`
       );
     }
 

open-sse/handlers/chatCore.ts

@@ -166,6 +166,11 @@ import {
   runWithCasGuard,
 } from "../services/tokenRefresh.ts";
 import { createRequestLogger } from "../utils/requestLogger.ts";
+import {
+  redactVideoTranscriptSensitiveText,
+  resolveVideoTranscriptLogSensitivity,
+} from "@/lib/guardrails/videoTranscriptLogRedaction";
+import { createExecutorRetentionLog } from "./chatCore/executorRetentionLog.ts";
 import { createPreparedRequestLogger, runWithCapture } from "../utils/providerRequestLogging.ts";
 import { summarizeToolSources } from "../utils/toolSources.ts";
 import { applyResponsesPreviousResponseIdPolicy } from "../utils/responsesStatePolicy.ts";
@@ -524,7 +529,22 @@ export async function handleChatCore({
   skipResourcePressureGuard = false,
   reasoningTransportFallback = "drop",
   managedLease = null,
+  videoTranscriptSensitive = false,
+  videoTranscriptDescriptionFingerprints = [],
 }) {
+  videoTranscriptSensitive =
+    videoTranscriptSensitive ||
+    resolveVideoTranscriptLogSensitivity({
+      rawRequestBody: clientRawRequest?.body,
+      processedBody: body,
+      trustedDescriptionFingerprints: videoTranscriptDescriptionFingerprints,
+    });
+  const retainedErrorTextForLog = (error: unknown): string =>
+    redactVideoTranscriptSensitiveText(
+      error instanceof Error ? error.message : String(error),
+      videoTranscriptSensitive
+    );
+  const retainedExecutorLog = createExecutorRetentionLog(log, videoTranscriptSensitive);
   let { provider, model, extendedContext } = modelInfo;
   const resilienceSettings = resolveResilienceSettings(cachedSettings);
   if (!skipResourcePressureGuard) {
@@ -637,6 +657,7 @@ export async function handleChatCore({
     apiKeyInfo,
     headers: clientRawRequest?.headers,
     log,
+    videoTranscriptSensitive,
   });
   if (pluginGate.blocked === true) {
     return {
@@ -721,6 +742,7 @@ export async function handleChatCore({
     effectiveServiceTier,
     startTime,
     log,
+    videoTranscriptSensitive,
   });
   if (idempotencyHit) {
     return idempotencyHit;
@@ -902,6 +924,8 @@ export async function handleChatCore({
       stage: "registered",
       correlationId,
       sessionTag: conversationId || null,
+      videoTranscriptSensitive,
+      videoTranscriptDescriptionFingerprints,
     }) || generateRequestId();
 
   // Initialize rate limit settings from persisted DB (once, lazy)
@@ -1034,6 +1058,8 @@ export async function handleChatCore({
       noLogEnabled,
       correlationId,
       modelPinned,
+      videoTranscriptSensitive,
+      videoTranscriptDescriptionFingerprints,
       // Resolved conversationId (open-sse/services/conversationTracker.ts) wins when
       // present — it's populated for every request now, not just ones where the
       // client explicitly sent x-omniroute-session-id. The raw header remains a
@@ -1156,8 +1182,17 @@ export async function handleChatCore({
     model,
     provider: provider || undefined,
     connectionId: connectionId || credentials?.connectionId || undefined,
+    videoTranscriptSensitive,
+    videoTranscriptDescriptionFingerprints,
   });
-  const pendingScope = { id: pendingRequestId, model, provider, connectionId: pendingConnId };
+  const pendingScope = {
+    id: pendingRequestId,
+    model,
+    provider,
+    connectionId: pendingConnId,
+    videoTranscriptSensitive,
+    videoTranscriptDescriptionFingerprints,
+  };
   const providerRequestCapture = createPreparedRequestLogger(reqLogger, pendingScope);
   // 0. Log client raw request (before format conversion)
   if (clientRawRequest) {
@@ -1203,6 +1238,7 @@ export async function handleChatCore({
     apiKeyId: apiKeyInfo?.id ?? undefined,
     cacheDefaultMode: (apiKeyInfo as { cacheDefaultMode?: "legacy" | "bypass" } | null)
       ?.cacheDefaultMode,
+    videoTranscriptSensitive,
   });
   if (cacheHit) {
     return cacheHit;
@@ -1482,8 +1518,7 @@ export async function handleChatCore({
         } catch (err) {
           log?.debug?.(
             "COMPRESSION",
-            "Combo compression override lookup skipped: " +
+            "Combo compression override lookup skipped: " + retainedErrorTextForLog(err)
-              (err instanceof Error ? err.message : String(err))
           );
         }
       }
@@ -1492,10 +1527,7 @@ export async function handleChatCore({
         const { listCompressionCombos } = await import("../../src/lib/db/compressionCombos.ts");
         namedCombos = buildNamedComboLookup(listCompressionCombos());
       } catch (err) {
-        log?.debug?.(
+        log?.debug?.("COMPRESSION", "Named combos load skipped: " + retainedErrorTextForLog(err));
-          "COMPRESSION",
-          "Named combos load skipped: " + (err instanceof Error ? err.message : String(err))
-        );
       }
       // Phase 3: per-request override. Unknown values fall through in the resolver (never error).
       const compressionHeader = resolveCompressionHeader(clientRawRequest?.headers ?? null);
@@ -1543,8 +1575,7 @@ export async function handleChatCore({
         } catch (err) {
           log?.debug?.(
             "COMPRESSION",
-            "Default compression combo lookup skipped: " +
+            "Default compression combo lookup skipped: " + retainedErrorTextForLog(err)
-              (err instanceof Error ? err.message : String(err))
           );
         }
       }
@@ -1586,10 +1617,7 @@ export async function handleChatCore({
             }
           }
         } catch (err) {
-          log?.debug?.(
+          log?.debug?.("COMPRESSION", "Output styles skipped: " + retainedErrorTextForLog(err));
-            "COMPRESSION",
-            "Output styles skipped: " + (err instanceof Error ? err.message : String(err))
-          );
         }
       }
       const compressionInputBody = body as Record<string, unknown>;
@@ -1899,8 +1927,7 @@ export async function handleChatCore({
     } catch (err) {
       log?.warn?.(
         "COMPRESSION",
-        "Compression pipeline error (non-fatal): " +
+        "Compression pipeline error (non-fatal): " + retainedErrorTextForLog(err)
-          (err instanceof Error ? err.message : String(err))
       );
     }
     // --- End Modular Compression Pipeline ---
@@ -1954,7 +1981,10 @@ export async function handleChatCore({
           `Combo context limit: ${resolved.limit} (source=${resolved.source})`
         );
       } catch (err) {
-        log?.warn?.("CONTEXT", "Failed to resolve combo limits for compression: " + err);
+        log?.warn?.(
+          "CONTEXT",
+          "Failed to resolve combo limits for compression: " + retainedErrorTextForLog(err)
+        );
       }
     }
 
@@ -2432,13 +2462,21 @@ export async function handleChatCore({
     try {
       const { runOnError } = await import("@/lib/plugins/hooks");
       await runOnError(
-        { requestId: traceId, body, model, provider, apiKeyInfo, metadata: {} },
+        {
+          requestId: traceId,
+          body,
+          model,
+          provider,
+          apiKeyInfo,
+          metadata: {},
+          videoTranscriptSensitive,
+        },
         error instanceof Error ? error : new Error(String(error))
       );
     } catch (pluginErr) {
       log?.debug?.(
         "PLUGIN",
-        `onError hook error (non-fatal): ${pluginErr instanceof Error ? pluginErr.message : String(pluginErr)}`
+        `onError hook error (non-fatal): ${retainedErrorTextForLog(pluginErr)}`
       );
     }
 
@@ -2450,7 +2488,13 @@ export async function handleChatCore({
     const message = error?.message || "Invalid request";
     const errorType = typeof error?.errorType === "string" ? error.errorType : null;
 
-    log?.warn?.("TRANSLATE", `Request translation failed: ${message}`);
+    log?.warn?.(
+      "TRANSLATE",
+      `Request translation failed: ${redactVideoTranscriptSensitiveText(
+        message,
+        videoTranscriptSensitive
+      )}`
+    );
 
     if (errorType) {
       trackPendingRequest(model, provider, connectionId, false);
@@ -2544,8 +2588,7 @@ export async function handleChatCore({
       // must never turn an otherwise valid translated request into a 500.
       log?.warn?.(
         "COMPRESSION",
-        "Post-translation OmniGlyph skipped: " +
+        "Post-translation OmniGlyph skipped: " + retainedErrorTextForLog(error)
-          (error instanceof Error ? error.message : String(error))
       );
     }
   }
@@ -2839,7 +2882,7 @@ export async function handleChatCore({
   const resolveExecutorWithProxy = (prov: string) =>
     resolveExecutorWithProxyFor(
       prov,
-      log,
+      retainedExecutorLog,
       (credentials?.providerSpecificData as Record<string, unknown> | null | undefined) ?? null
     );
 
@@ -2861,7 +2904,7 @@ export async function handleChatCore({
       }).catch((err: unknown): EnforceDecision => {
         log?.warn?.(
           "QUOTA_SHARE",
-          `enforceQuotaShare failed; fail-open: ${err instanceof Error ? err.message : String(err)}`
+          `enforceQuotaShare failed; fail-open: ${retainedErrorTextForLog(err)}`
         );
         return { kind: "allow" as const };
       });
@@ -2903,7 +2946,7 @@ export async function handleChatCore({
       // Outer fail-open guard — should not be reached (inner .catch covers it)
       log?.warn?.(
         "QUOTA_SHARE",
-        `[quotaShare] enforceQuotaShare unexpected error; fail-open: ${err instanceof Error ? err.message : String(err)}`
+        `[quotaShare] enforceQuotaShare unexpected error; fail-open: ${retainedErrorTextForLog(err)}`
       );
     }
   }
@@ -2915,7 +2958,7 @@ export async function handleChatCore({
     } catch (err) {
       log?.warn?.(
         "QUOTA_SHARE",
-        `[quotaShare] could not set soft penalty on candidate: ${err instanceof Error ? err.message : String(err)}`
+        `[quotaShare] could not set soft penalty on candidate: ${retainedErrorTextForLog(err)}`
       );
     }
   }
@@ -2949,6 +2992,8 @@ export async function handleChatCore({
   let onPipelineStreamError: streamFailure.PipelineStreamErrorHandler | null = null;
   let onClientDisconnectFinalize:
     ((event: { reason: string; duration: number }) => boolean) | null = null;
+  const redactStreamDiagnosticsForLog =
+    videoTranscriptSensitive || reqLogger.isVideoTranscriptSensitive();
 
   // Create stream controller for disconnect detection
   const streamController = createStreamController({
@@ -2980,6 +3025,7 @@ export async function handleChatCore({
     clientAbortSignal: clientRawRequest?.signal,
     allowCompletedToolHandoffGrace: isCodexResponsesEcho,
     clientDisconnectGracePeriodMs: STREAM_DISCONNECT_GRACE_PERIOD_MS,
+    redactStreamDiagnosticsForLog,
   });
 
   const dedupRequestBody = { ...translatedBody, model: `${provider}/${model}`, stream };
@@ -3112,7 +3158,7 @@ export async function handleChatCore({
                       execCreds?.providerSpecificData
                     ),
                     signal: streamController.signal,
-                    log,
+                    log: retainedExecutorLog,
                     execute: (signal) =>
                       runWithCapture(providerRequestCapture, () =>
                         executor.execute({
@@ -3121,7 +3167,7 @@ export async function handleChatCore({
                           stream: upstreamStream,
                           credentials: execCreds,
                           signal,
-                          log,
+                          log: retainedExecutorLog,
                           extendedContext,
                           upstreamExtraHeaders: buildUpstreamHeadersForExecute(modelToCall),
                           clientHeaders: buildExecutorClientHeaders(
@@ -3129,6 +3175,7 @@ export async function handleChatCore({
                             userAgent
                           ),
                           clientResponseFormat,
+                          videoTranscriptSensitive,
                           onCredentialsRefreshed,
                           skipUpstreamRetry,
                           contextEditing: { enabled: contextEditingEnabled },
@@ -3163,7 +3210,7 @@ export async function handleChatCore({
                     invalidateCodexQuotaCache(String(attemptConnectionId));
                   }
                 } catch (err) {
-                  const errMessage = err instanceof Error ? err.message : String(err);
+                  const errMessage = retainedErrorTextForLog(err);
                   log?.debug?.("CODEX", `Failed to persist codex quota state: ${errMessage}`);
                 }
               }
@@ -3428,7 +3475,7 @@ export async function handleChatCore({
                           execCreds?.providerSpecificData
                         ),
                         signal: streamController.signal,
-                        log,
+                        log: retainedExecutorLog,
                         execute: (signal) =>
                           runWithCapture(providerRequestCapture, () =>
                             executor.execute({
@@ -3437,7 +3484,7 @@ export async function handleChatCore({
                               stream: upstreamStream,
                               credentials: execCreds,
                               signal,
-                              log,
+                              log: retainedExecutorLog,
                               extendedContext,
                               upstreamExtraHeaders: buildUpstreamHeadersForExecute(modelToCall),
                               clientHeaders: buildExecutorClientHeaders(
@@ -3445,6 +3492,7 @@ export async function handleChatCore({
                                 userAgent
                               ),
                               clientResponseFormat,
+                              videoTranscriptSensitive,
                               onCredentialsRefreshed,
                               skipUpstreamRetry,
                               contextEditing: { enabled: contextEditingEnabled },
@@ -3677,7 +3725,9 @@ export async function handleChatCore({
     } catch (err) {
       // Fail-open at Tier 2: Tier 1 already enforced the model/global limit pre-dispatch.
       // A transient counter read error here must not break an otherwise-valid request.
-      log?.warn?.("TOKEN_LIMIT", "Tier 2 token-limit check failed; allowing request", { err });
+      log?.warn?.("TOKEN_LIMIT", "Tier 2 token-limit check failed; allowing request", {
+        error: retainedErrorTextForLog(err),
+      });
     }
   }
 
@@ -3696,7 +3746,9 @@ export async function handleChatCore({
         );
       }
     } catch (err) {
-      log?.warn?.("GEMINI_RATE_LIMIT", "Pre-dispatch TPM check failed; allowing request", { err });
+      log?.warn?.("GEMINI_RATE_LIMIT", "Pre-dispatch TPM check failed; allowing request", {
+        error: retainedErrorTextForLog(err),
+      });
     }
   }
 
@@ -3846,7 +3898,12 @@ export async function handleChatCore({
       failureStatus,
       upstreamErrorCode || (error instanceof Error && error.name ? error.name : "upstream_error")
     );
-    console.log(`${COLORS.red}[ERROR] ${failureMessage}${COLORS.reset}`);
+    console.log(
+      `${COLORS.red}[ERROR] ${redactVideoTranscriptSensitiveText(
+        failureMessage,
+        videoTranscriptSensitive
+      )}${COLORS.reset}`
+    );
     if (stream && upstreamErrorCode) {
       const result = createStreamingErrorResult(
         failureStatus,
@@ -3978,11 +4035,12 @@ export async function handleChatCore({
               stream: upstreamStream,
               credentials: getExecutionCredentials(),
               signal: streamController.signal,
-              log,
+              log: retainedExecutorLog,
               extendedContext,
               upstreamExtraHeaders: buildUpstreamHeadersForExecute(retryModelId),
               clientHeaders: buildExecutorClientHeaders(clientRawRequest?.headers, userAgent),
               clientResponseFormat,
+              videoTranscriptSensitive,
               onCredentialsRefreshed,
               skipUpstreamRetry: isCombo,
               contextEditing: { enabled: contextEditingEnabled },
@@ -4012,9 +4070,13 @@ export async function handleChatCore({
         // executor throw). Don't swallow — the operator-visible signal "the user
         // saw 401 even though auth was actually fixed" is much more confusing
         // than the original 401 alone. Surface at error level with sanitization.
+        const retainedRetryError = redactVideoTranscriptSensitiveText(
+          sanitizeErrorMessage(retryErr),
+          videoTranscriptSensitive
+        );
         log?.error?.(
           "TOKEN",
-          `${provider?.toUpperCase()} | retry after refresh failed: ${sanitizeErrorMessage(retryErr)}`
+          `${provider?.toUpperCase()} | retry after refresh failed: ${retainedRetryError}`
         );
       }
     } else {
@@ -4131,6 +4193,11 @@ export async function handleChatCore({
 
     if (signatureRecovery.succeeded) break providerFailure;
 
+    const retainedProviderMessage = redactVideoTranscriptSensitiveText(
+      message,
+      videoTranscriptSensitive
+    );
+
     // #10281 — tiny-budget reasoning probes (e.g. Claude Code's `/model` check
     // sends `max_tokens: 1`): the model burns the whole budget on thinking, and
     // some upstreams (e.g. api.cline.bot for deepseek-v4-flash) answer the empty
@@ -4153,7 +4220,7 @@ export async function handleChatCore({
       });
       log?.warn?.(
         "PROBE",
-        `Reasoning probe (max_tokens < ${REASONING_BUFFER_MIN_TRIGGER}) answered with truncated 200 — upstream reported "${message}"`
+        `Reasoning probe (max_tokens < ${REASONING_BUFFER_MIN_TRIGGER}) answered with truncated 200 — upstream reported "${retainedProviderMessage}"`
       );
       break providerFailure;
     }
@@ -4182,7 +4249,7 @@ export async function handleChatCore({
               {
                 testStatus: "banned",
                 isActive: false,
-                lastError: message,
+                lastError: retainedProviderMessage,
                 lastErrorType: errorType,
                 errorCode: String(statusCode),
               },
@@ -4213,7 +4280,7 @@ export async function handleChatCore({
           ) {
             await updateProviderConnection(errorConnectionId, {
               lastErrorType: errorType,
-              lastError: message,
+              lastError: retainedProviderMessage,
               errorCode: statusCode,
             });
             console.warn(
@@ -4226,7 +4293,7 @@ export async function handleChatCore({
               {
                 testStatus: "deactivated",
                 isActive: false,
-                lastError: message,
+                lastError: retainedProviderMessage,
                 lastErrorType: errorType,
                 errorCode: String(statusCode),
               },
@@ -4250,7 +4317,7 @@ export async function handleChatCore({
                 errorConnectionId,
                 {
                   testStatus: "credits_exhausted",
-                  lastError: message,
+                  lastError: retainedProviderMessage,
                   lastErrorType: errorType,
                   errorCode: String(statusCode),
                 },
@@ -4298,7 +4365,7 @@ export async function handleChatCore({
                   rateLimitedUntil: kimiRateLimitResetAt,
                   backoffLevel: 0,
                   lastErrorType: PROVIDER_ERROR_TYPES.RATE_LIMITED,
-                  lastError: message,
+                  lastError: retainedProviderMessage,
                   errorCode: statusCode,
                 });
                 console.warn(
@@ -4328,7 +4395,7 @@ export async function handleChatCore({
                   errorConnectionId,
                   {
                     testStatus: "credits_exhausted",
-                    lastError: message,
+                    lastError: retainedProviderMessage,
                     lastErrorType: errorType,
                     errorCode: String(statusCode),
                   },
@@ -4344,14 +4411,14 @@ export async function handleChatCore({
           // Normal 401 (token/session auth issue): keep account active for refresh/re-auth.
           await updateProviderConnection(errorConnectionId, {
             lastErrorType: errorType,
-            lastError: message,
+            lastError: retainedProviderMessage,
             errorCode: statusCode,
           });
         } else if (errorType === PROVIDER_ERROR_TYPES.OAUTH_INVALID_TOKEN) {
           // OAuth 401 with invalid credentials - token refresh can recover
           await updateProviderConnection(errorConnectionId, {
             lastErrorType: errorType,
-            lastError: message,
+            lastError: retainedProviderMessage,
             errorCode: statusCode,
           });
           console.warn(
@@ -4361,7 +4428,7 @@ export async function handleChatCore({
           // Cloud Code 403 with stale project: not a ban, keep account active.
           await updateProviderConnection(errorConnectionId, {
             lastErrorType: errorType,
-            lastError: message,
+            lastError: retainedProviderMessage,
             errorCode: statusCode,
           });
           console.warn(
@@ -4377,7 +4444,7 @@ export async function handleChatCore({
           const geoCooldownMs = COOLDOWN_MS.geoBlocked ?? 24 * 60 * 60 * 1000;
           await updateProviderConnection(errorConnectionId, {
             lastErrorType: errorType,
-            lastError: message,
+            lastError: retainedProviderMessage,
             errorCode: statusCode,
           });
           // T-PROBE: the 24h exclusion is a routing mutation — a probe must
@@ -4402,7 +4469,7 @@ export async function handleChatCore({
           const byopCooldownMs = COOLDOWN_MS.gcpProjectRequired ?? 24 * 60 * 60 * 1000;
           await updateProviderConnection(errorConnectionId, {
             lastErrorType: errorType,
-            lastError: message,
+            lastError: retainedProviderMessage,
             errorCode: statusCode,
           });
           try {
@@ -4448,7 +4515,13 @@ export async function handleChatCore({
     }).catch(() => {});
 
     const errMsg = formatProviderError(new Error(message), provider, model, statusCode);
-    console.log(`${COLORS.red}[ERROR] ${errMsg}${COLORS.reset}`);
+    const retainedErrMsg = formatProviderError(
+      new Error(retainedProviderMessage),
+      provider,
+      model,
+      statusCode
+    );
+    console.log(`${COLORS.red}[ERROR] ${retainedErrMsg}${COLORS.reset}`);
 
     // Log Antigravity retry time if available
     if (retryAfterMs && provider === "antigravity") {
@@ -4775,12 +4848,11 @@ export async function handleChatCore({
             }
           }
         } catch (retryErr) {
-          log?.warn?.(
+          const retainedRetryError = redactVideoTranscriptSensitiveText(
-            "RETRY",
+            retryErr instanceof Error ? retryErr.message : String(retryErr),
-            `clinepass retry failed: ${
+            videoTranscriptSensitive
-              retryErr instanceof Error ? retryErr.message : String(retryErr)
-            }`
           );
+          log?.warn?.("RETRY", `clinepass retry failed: ${retainedRetryError}`);
         }
       }
       if (envError) {
@@ -4999,6 +5071,7 @@ export async function handleChatCore({
         cacheReasoningFromAssistantMessage(msg, provider, model, {
           scope: reasoningCacheScope,
           historyMessages: Array.isArray(historyMessages) ? historyMessages : [],
+          videoTranscriptSensitive,
         });
       }
     } catch {
@@ -5050,12 +5123,18 @@ export async function handleChatCore({
     );
 
     if (memoryOwnerId && memorySettings?.enabled && memorySettings.maxTokens > 0) {
-      const requestMemoryText = extractMemoryTextFromRequestBody(body as Record<string, unknown>);
+      const requestMemoryText = extractMemoryTextFromRequestBody(
+        body as Record<string, unknown>,
+        videoTranscriptSensitive,
+        { trustedDescriptionFingerprints: videoTranscriptDescriptionFingerprints }
+      );
       if (requestMemoryText) {
         extractFacts(requestMemoryText, memoryOwnerId, pipelineSessionId);
       }
 
-      const memoryText = extractMemoryTextFromResponse(memoryExtractionResponse);
+      const memoryText = videoTranscriptSensitive
+        ? ""
+        : extractMemoryTextFromResponse(memoryExtractionResponse);
       if (memoryText) {
         extractFacts(memoryText, memoryOwnerId, pipelineSessionId);
       }
@@ -5095,6 +5174,7 @@ export async function handleChatCore({
       provider,
       responsePayloadFormat,
       clientResponseFormat,
+      videoTranscriptSensitive,
     });
     const postCallGuardrails = await guardrailRegistry.runPostCallHooks(
       translatedResponse,
@@ -5136,7 +5216,10 @@ export async function handleChatCore({
       }
       log?.warn?.(
         "GUARDRAIL",
-        `Response blocked by ${postCallGuardrails.guardrail || "guardrail"}: ${guardrailMessage}`
+        `Response blocked by ${postCallGuardrails.guardrail || "guardrail"}: ${redactVideoTranscriptSensitiveText(
+          guardrailMessage,
+          videoTranscriptSensitive
+        )}`
       );
       finalizePendingScope(pendingScope, {
         providerResponse: responseBody,
@@ -5239,6 +5322,7 @@ export async function handleChatCore({
       apiKeyId: apiKeyInfo?.id ?? undefined,
       usage,
       log,
+      videoTranscriptSensitive,
     });
 
     // ── Phase 9.2: Save for idempotency ──
@@ -5318,6 +5402,7 @@ export async function handleChatCore({
       apiKeyInfo,
       headers: clientRawRequest?.headers,
       response: { status: 200, data: translatedResponse },
+      videoTranscriptSensitive,
     });
 
     // Routing event (feedback foundation) — fire-and-forget, cheap.
@@ -5393,6 +5478,7 @@ export async function handleChatCore({
     provider,
     model,
     log,
+    redactUpstreamDiagnosticForLog: videoTranscriptSensitive,
   });
   if (streamReadiness.ok === false) {
     const { response: failureResponse, reason } = streamReadiness;
@@ -5524,6 +5610,7 @@ export async function handleChatCore({
           cacheReasoningFromAssistantMessage(msg, provider, model, {
             scope: reasoningCacheScope,
             historyMessages: Array.isArray(historyMessages) ? historyMessages : [],
+            videoTranscriptSensitive,
           });
         }
       } catch {
@@ -5556,6 +5643,7 @@ export async function handleChatCore({
       status: normalizedStreamStatus,
       error: streamError,
       errorCode: streamErrorCode,
+      videoTranscriptSensitive,
     });
 
     // Track cache token metrics for streaming responses
@@ -5678,14 +5766,20 @@ export async function handleChatCore({
       memorySettings.maxTokens > 0 &&
       streamStatus === 200
     ) {
-      const requestMemoryText = extractMemoryTextFromRequestBody(body as Record<string, unknown>);
+      const requestMemoryText = extractMemoryTextFromRequestBody(
+        body as Record<string, unknown>,
+        videoTranscriptSensitive,
+        { trustedDescriptionFingerprints: videoTranscriptDescriptionFingerprints }
+      );
       if (requestMemoryText) {
         extractFacts(requestMemoryText, memoryOwnerId, pipelineSessionId);
       }
 
-      const streamedMemoryText = extractMemoryTextFromResponse(
+      const streamedMemoryText = videoTranscriptSensitive
-        (streamResponseBody ?? null) as Record<string, unknown> | null
+        ? ""
-      );
+        : extractMemoryTextFromResponse(
+            (streamResponseBody ?? null) as Record<string, unknown> | null
+          );
       if (streamedMemoryText) {
         extractFacts(streamedMemoryText, memoryOwnerId, pipelineSessionId);
       }
@@ -5702,6 +5796,7 @@ export async function handleChatCore({
       apiKeyId: apiKeyInfo?.id ?? undefined,
       streamUsage,
       log,
+      videoTranscriptSensitive,
     });
 
     // Plugin onStreamComplete hook — fire-and-forget, fail-open (#9571)
@@ -5770,7 +5865,8 @@ export async function handleChatCore({
       // openai-responses → openai translation still wants the namespace identity
       // map for #7936-style round-trip closure when the client also speaks
       // Responses (Codex CLI).
-      requestToolIdentityMap
+      requestToolIdentityMap,
+      redactStreamDiagnosticsForLog
     );
   } else if (needsTranslation(targetFormat, clientResponseFormat)) {
     // Standard translation for other providers
@@ -5800,7 +5896,8 @@ export async function handleChatCore({
         clientResponseFormat,
       }),
       customToolNames,
-      requestToolIdentityMap
+      requestToolIdentityMap,
+      redactStreamDiagnosticsForLog
     );
   } else {
     log?.debug?.("STREAM", `Standard passthrough mode`);
@@ -5815,7 +5912,8 @@ export async function handleChatCore({
       apiKeyInfo,
       handleStreamFailure,
       clientResponseFormat,
-      requestToolIdentityMap
+      requestToolIdentityMap,
+      redactStreamDiagnosticsForLog
     );
   }
 
@@ -5827,6 +5925,7 @@ export async function handleChatCore({
     clientRawRequestHeaders: clientRawRequest?.headers,
     clientResponseFormat,
     echoModel,
+    redactStreamDiagnosticsForLog,
     responseHeaders,
   });
 
@@ -5842,6 +5941,7 @@ export async function handleChatCore({
     apiKeyInfo,
     headers: clientRawRequest?.headers,
     response: { status: 200, streamed: true },
+    videoTranscriptSensitive,
   });
 
   return {

open-sse/handlers/chatCore/attemptLogging.ts

@@ -15,6 +15,10 @@ import { logAuditEvent } from "@/lib/compliance";
 import { emit } from "@/lib/events/eventBus";
 import type { RequestCompletedPayload, RequestFailedPayload } from "@/lib/events/types";
 import { saveCallLog } from "@/lib/usageDb";
+import {
+  VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+  VIDEO_TRANSCRIPT_REDACTION_PIPELINE_KEY,
+} from "@/lib/guardrails/videoTranscriptLogRedaction";
 import { FORMATS } from "../../translator/formats.ts";
 import { takeEarlyKeepaliveBytes } from "../../utils/earlyKeepaliveByteBuffer.ts";
 import { cloneBoundedChatLogPayload, truncateForLog } from "./logTruncation.ts";
@@ -68,7 +72,13 @@ export type PersistAttemptLogsContext = {
   model: string | null | undefined;
   skillRequestId: string;
   detailedLoggingEnabled: boolean;
-  reqLogger: { getPipelinePayloads?: () => Record<string, unknown> | undefined } | null | undefined;
+  reqLogger:
+    | {
+        getPipelinePayloads?: () => Record<string, unknown> | undefined;
+        isVideoTranscriptSensitive?: () => boolean;
+      }
+    | null
+    | undefined;
   pendingRequestId: unknown;
   clientRawRequest: { endpoint?: string } | null | undefined;
   requestedModel: unknown;
@@ -89,12 +99,34 @@ export type PersistAttemptLogsContext = {
    * explicitly present (never synthesized from skillRequestId) — persisted as call_logs.session_tag
    * for per-session cost attribution. */
   sessionTag?: string | null;
+  /** Trusted request state derived from the Video Bridge guardrail result. */
+  videoTranscriptSensitive?: boolean;
+  /** Exact SHA-256 identities of transcript descriptions generated by the guardrail. */
+  videoTranscriptDescriptionFingerprints?: readonly string[];
 };
 
 function toConnectionId(value: unknown): string | null {
   return typeof value === "string" && value.trim().length > 0 ? value.trim() : null;
 }
 
+function omitSensitivePipelineResponse(value: unknown): Record<string, unknown> {
+  const record =
+    value && typeof value === "object" && !Array.isArray(value)
+      ? (value as Record<string, unknown>)
+      : {};
+  return {
+    ...(typeof record.timestamp === "string" ? { timestamp: record.timestamp } : {}),
+    ...(typeof record.status === "number" && Number.isFinite(record.status)
+      ? { status: record.status }
+      : {}),
+    ...(record.headers !== undefined ? { headers: VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER } : {}),
+    ...(record.statusText !== undefined
+      ? { statusText: VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER }
+      : {}),
+    body: VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+  };
+}
+
 function buildAccountRotationMeta(
   provider: string | null | undefined,
   initialConnectionId: string | null,
@@ -204,6 +236,8 @@ export function persistAttemptLogs(args: PersistAttemptLogsArgs, ctx: PersistAtt
     correlationId,
     modelPinned,
     sessionTag,
+    videoTranscriptSensitive,
+    videoTranscriptDescriptionFingerprints,
   } = ctx;
   const initialConnectionId = toConnectionId(connectionId);
   const finalConnectionId = toConnectionId(credentials?.connectionId) || initialConnectionId;
@@ -212,8 +246,21 @@ export function persistAttemptLogs(args: PersistAttemptLogsArgs, ctx: PersistAtt
     initialConnectionId,
     finalConnectionId
   );
+  const transcriptSensitive =
+    videoTranscriptSensitive === true || reqLogger?.isVideoTranscriptSensitive?.() === true;
+  const descriptionLogContext = {
+    trustedDescriptionFingerprints: videoTranscriptDescriptionFingerprints ?? [],
+  };
 
-  const providerWarnings = extractProviderWarnings(providerResponse, clientResponse, responseBody);
+  const detectedProviderWarnings = extractProviderWarnings(
+    providerResponse,
+    clientResponse,
+    responseBody
+  );
+  const providerWarnings =
+    transcriptSensitive && detectedProviderWarnings.length > 0
+      ? [VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER]
+      : detectedProviderWarnings;
   if (providerWarnings.length > 0) {
     logAuditEvent({
       action: "provider.warning",
@@ -227,6 +274,7 @@ export function persistAttemptLogs(args: PersistAttemptLogsArgs, ctx: PersistAtt
         model,
         connectionId: finalConnectionId,
         httpStatus: status,
+        warningCount: detectedProviderWarnings.length,
         warnings: providerWarnings,
       },
     });
@@ -273,8 +321,51 @@ export function persistAttemptLogs(args: PersistAttemptLogsArgs, ctx: PersistAtt
         };
       }
     }
+    if (transcriptSensitive) {
+      pipelinePayloads[VIDEO_TRANSCRIPT_REDACTION_PIPELINE_KEY] = true;
+      for (const requestStage of [
+        "clientRawRequest",
+        "openaiRequest",
+        "providerRequest",
+      ] as const) {
+        if (pipelinePayloads[requestStage] !== undefined) {
+          pipelinePayloads[requestStage] = cloneBoundedChatLogPayload(
+            pipelinePayloads[requestStage],
+            0,
+            requestStage !== "clientRawRequest" ? descriptionLogContext : {}
+          ) as Record<string, unknown>;
+        }
+      }
+      if (pipelinePayloads.providerResponse !== undefined || providerResponse !== undefined) {
+        pipelinePayloads.providerResponse = omitSensitivePipelineResponse(
+          pipelinePayloads.providerResponse ?? providerResponse
+        );
+      }
+      if (pipelinePayloads.clientResponse !== undefined || clientResponse !== undefined) {
+        pipelinePayloads.clientResponse = omitSensitivePipelineResponse(
+          pipelinePayloads.clientResponse ?? clientResponse
+        );
+      }
+      if (pipelinePayloads.error !== undefined || error) {
+        const errorRecord =
+          pipelinePayloads.error && typeof pipelinePayloads.error === "object"
+            ? pipelinePayloads.error
+            : {};
+        pipelinePayloads.error = {
+          ...(typeof errorRecord.timestamp === "string"
+            ? { timestamp: errorRecord.timestamp }
+            : {}),
+          message: VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+        };
+      }
+      delete pipelinePayloads.streamChunks;
+    }
   }
 
+  const responseBodyForLog = transcriptSensitive
+    ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER
+    : responseBody;
+
   saveCallLog({
     id: pendingRequestId,
     method: "POST",
@@ -290,10 +381,12 @@ export function persistAttemptLogs(args: PersistAttemptLogsArgs, ctx: PersistAtt
       attachLogMeta(truncateForLog(body as Record<string, unknown>), {
         ...accountRotationMeta,
         claudePromptCache: claudeCacheMeta,
-      })
+      }),
+      0,
+      descriptionLogContext
     ),
     responseBody: cloneBoundedChatLogPayload(
-      attachLogMeta(truncateForLog(responseBody as Record<string, unknown>), {
+      attachLogMeta(truncateForLog(responseBodyForLog as Record<string, unknown>), {
         ...accountRotationMeta,
         claudePromptCache: claudeCacheMeta
           ? {
@@ -305,7 +398,7 @@ export function persistAttemptLogs(args: PersistAttemptLogsArgs, ctx: PersistAtt
         claudePromptCacheUsage: claudeCacheUsageMeta,
       })
     ),
-    error: error || null,
+    error: transcriptSensitive && error ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER : error || null,
     sourceFormat,
     targetFormat,
     comboName,
@@ -332,7 +425,7 @@ export function persistAttemptLogs(args: PersistAttemptLogsArgs, ctx: PersistAtt
     const lifecycle = resolveRequestLifecycleEvent({
       traceId,
       status,
-      error,
+      error: transcriptSensitive && error ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER : error,
       model,
       provider,
       comboName,

open-sse/handlers/chatCore/executorProxy.ts

@@ -17,8 +17,10 @@
  */
 
 import { getExecutor } from "../../executors/index.ts";
+import type { ExecuteInput } from "../../executors/base.ts";
 import { isCliproxyapiDeepModeEnabled } from "../../executors/cliproxyapi.ts";
 import { isDarioDeepModeEnabled } from "../../executors/dario.ts";
+import { redactVideoTranscriptSensitiveText } from "../../../src/lib/guardrails/videoTranscriptLogRedaction.ts";
 import { getCachedSettings } from "@/lib/db/readCache";
 import { getUpstreamProxyConfigCached } from "./comboContextCache.ts";
 import type { FallbackBackend } from "@/lib/db/upstreamProxy";
@@ -48,6 +50,11 @@ function parseFallbackCodes(raw: unknown): number[] | null {
   return parsed.length > 0 ? parsed : null;
 }
 
+function retainExecutorDiagnostic(error: unknown, input: ExecuteInput): string {
+  const message = error instanceof Error ? error.message : String(error);
+  return redactVideoTranscriptSensitiveText(message, input.videoTranscriptSensitive === true);
+}
+
 /**
  * Reads the CLIProxyAPI-related settings shared by both the direct
  * `mode: "cliproxyapi"` passthrough leg and the `mode: "fallback"` retry leg:
@@ -155,25 +162,20 @@ export async function resolveExecutorWithProxy(
   const isRetryableStatus = (s: number) => fallbackCodes.includes(s) || s === 0;
 
   const wrapper = Object.create(nativeExec);
-  wrapper.execute = async (input: {
+  wrapper.execute = async (input: ExecuteInput) => {
-    model: string;
-    body: unknown;
-    stream: boolean;
-    credentials: unknown;
-    signal?: AbortSignal | null;
-    log?: unknown;
-    upstreamExtraHeaders?: Record<string, string> | null;
-  }) => {
     let result;
     try {
       result = await nativeExec.execute(input);
     } catch (err) {
-      const errMsg = err instanceof Error ? err.message : String(err);
+      const errMsg = retainExecutorDiagnostic(err, input);
-      log?.info?.("UPSTREAM_PROXY", `${prov} native error (${errMsg}), retrying via ${backendLabel}`);
+      log?.info?.(
+        "UPSTREAM_PROXY",
+        `${prov} native error (${errMsg}), retrying via ${backendLabel}`
+      );
       try {
         return await proxyExec.execute(input);
       } catch (proxyErr) {
-        const proxyMsg = proxyErr instanceof Error ? proxyErr.message : String(proxyErr);
+        const proxyMsg = retainExecutorDiagnostic(proxyErr, input);
         log?.error?.("UPSTREAM_PROXY", `${prov} ${backendLabel} fallback also failed: ${proxyMsg}`);
         throw proxyErr;
       }
@@ -189,7 +191,7 @@ export async function resolveExecutorWithProxy(
     try {
       return await proxyExec.execute(input);
     } catch (proxyErr) {
-      const proxyMsg = proxyErr instanceof Error ? proxyErr.message : String(proxyErr);
+      const proxyMsg = retainExecutorDiagnostic(proxyErr, input);
       log?.error?.("UPSTREAM_PROXY", `${prov} ${backendLabel} fallback also failed: ${proxyMsg}`);
       throw proxyErr;
     }

open-sse/handlers/chatCore/executorRetentionLog.ts

@@ -0,0 +1,36 @@
+import { VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER } from "@/lib/guardrails/videoTranscriptLogRedaction";
+
+import type { ExecutorLog } from "../../executors/base.ts";
+
+type ExecutorLogMethod = NonNullable<ExecutorLog["debug"]>;
+
+function retainMethod(
+  owner: ExecutorLog,
+  method: ExecutorLogMethod | undefined
+): ExecutorLogMethod | undefined {
+  if (!method) return undefined;
+
+  return (tag) => {
+    method.call(owner, tag, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER);
+  };
+}
+
+/**
+ * Executor diagnostics can contain arbitrary provider text, including request echoes.
+ * Keep the original logger for ordinary requests, but give every executor a request-scoped
+ * fail-closed view when the Video Bridge marked the request sensitive. Tags remain useful;
+ * messages and optional structured metadata are replaced/dropped only at the retention seam.
+ */
+export function createExecutorRetentionLog(
+  log: ExecutorLog | null | undefined,
+  videoTranscriptSensitive: boolean
+): ExecutorLog | null | undefined {
+  if (!log || !videoTranscriptSensitive) return log;
+
+  return {
+    debug: retainMethod(log, log.debug),
+    info: retainMethod(log, log.info),
+    warn: retainMethod(log, log.warn),
+    error: retainMethod(log, log.error),
+  };
+}

open-sse/handlers/chatCore/idempotency.ts

@@ -124,6 +124,7 @@ export async function checkIdempotencyCache({
   effectiveServiceTier,
   startTime,
   log,
+  videoTranscriptSensitive,
 }: {
   clientRawRequest: IdempotencyRequest;
   provider: string;
@@ -132,7 +133,9 @@ export async function checkIdempotencyCache({
   effectiveServiceTier: EffectiveServiceTier | null | undefined;
   startTime: number;
   log: LoggerLike;
+  videoTranscriptSensitive: boolean;
 }): Promise<{ hit: { success: true; response: Response } | null; idempotencyKey: string | null }> {
+  if (videoTranscriptSensitive) return { hit: null, idempotencyKey: null };
   // NEXA fusion-idempotency fix: namespace the raw header key (see composeIdempotencyKey).
   const rawIdempotencyKey = getIdempotencyKey(clientRawRequest?.headers);
   const idempotencyKey = composeIdempotencyKey({

open-sse/handlers/chatCore/logTruncation.ts

@@ -5,6 +5,10 @@ import {
   getChatLogMaxObjectKeys,
   getChatLogMaxBodyBytes,
 } from "@/lib/logEnv";
+import {
+  omitVideoTranscriptForLog,
+  type VideoTranscriptLogContext,
+} from "@/lib/guardrails/videoTranscriptLogRedaction";
 import { estimateSizeFast } from "../../utils/estimateSize.ts";
 
 export const MEMORY_EXTRACTION_TEXT_LIMIT = 64 * 1024;
@@ -22,7 +26,7 @@ export function truncateChatLogText(value: string): string {
   return `${head}\n[...truncated ${value.length - limit} chars...]\n${tail}`;
 }
 
-export function cloneBoundedChatLogPayload(value: unknown, depth = 0): unknown {
+function cloneBoundedChatLogPayloadValue(value: unknown, depth = 0): unknown {
   if (value === null || value === undefined) return value;
   if (typeof value === "string") return truncateChatLogText(value);
   if (typeof value !== "object") return value;
@@ -32,7 +36,7 @@ export function cloneBoundedChatLogPayload(value: unknown, depth = 0): unknown {
 
   if (Array.isArray(value)) {
     const retained = value.length > maxTailItems ? value.slice(-maxTailItems) : value;
-    const cloned = retained.map((item) => cloneBoundedChatLogPayload(item, depth + 1));
+    const cloned = retained.map((item) => cloneBoundedChatLogPayloadValue(item, depth + 1));
     if (value.length > maxTailItems) {
       return [
         {
@@ -47,10 +51,11 @@ export function cloneBoundedChatLogPayload(value: unknown, depth = 0): unknown {
   }
 
   const result: Record<string, unknown> = {};
-  const entries = Object.entries(value as Record<string, unknown>);
+  const record = value as Record<string, unknown>;
+  const entries = Object.entries(record);
   const maxKeys = getChatLogMaxObjectKeys();
   for (const [key, item] of maxKeys > 0 ? entries.slice(0, maxKeys) : entries) {
-    result[key] = cloneBoundedChatLogPayload(item, depth + 1);
+    result[key] = cloneBoundedChatLogPayloadValue(item, depth + 1);
   }
   if (maxKeys > 0 && entries.length > maxKeys) {
     result._omniroute_truncated_keys = entries.length - maxKeys;
@@ -58,6 +63,16 @@ export function cloneBoundedChatLogPayload(value: unknown, depth = 0): unknown {
   return result;
 }
 
+export function cloneBoundedChatLogPayload(
+  value: unknown,
+  depth = 0,
+  descriptionContext: VideoTranscriptLogContext = {}
+): unknown {
+  const transcriptSafeValue =
+    depth === 0 ? omitVideoTranscriptForLog(value, descriptionContext) : value;
+  return cloneBoundedChatLogPayloadValue(transcriptSafeValue, depth);
+}
+
 /**
  * Truncate a large object for logging. If its JSON representation exceeds
  * getChatLogMaxBodyBytes() (default 1MB; CHAT_LOG_MAX_BODY_KB env override),

open-sse/handlers/chatCore/memoryExtraction.ts

@@ -1,5 +1,39 @@
+import {
+  containsVideoTranscriptForLog,
+  omitVideoTranscriptDerivedTextForMemory,
+  type VideoTranscriptLogContext,
+} from "../../../src/lib/guardrails/videoTranscriptLogRedaction.ts";
 import { capMemoryExtractionText, MEMORY_EXTRACTION_TEXT_LIMIT } from "./logTruncation.ts";
 
+function normalizeMemoryInputText(value: unknown, context: VideoTranscriptLogContext = {}): string {
+  if (typeof value !== "string") return "";
+  return omitVideoTranscriptDerivedTextForMemory(value, context).trim();
+}
+
+function extractMemoryTextPart(
+  part: Record<string, unknown>,
+  transcriptSensitive: boolean,
+  context: VideoTranscriptLogContext
+): string {
+  try {
+    const rawText = typeof part?.text === "string" ? part.text : "";
+    if (!rawText) return "";
+
+    const retainedText = normalizeMemoryInputText(rawText, context);
+    if (!retainedText) return "";
+    if (
+      transcriptSensitive &&
+      containsVideoTranscriptForLog(part, context) &&
+      retainedText === rawText.trim()
+    ) {
+      return "";
+    }
+    return retainedText;
+  } catch {
+    return "";
+  }
+}
+
 export function extractMemoryTextFromResponse(
   response: Record<string, unknown> | null | undefined
 ): string {
@@ -29,9 +63,16 @@ export function extractMemoryTextFromResponse(
 }
 
 export function extractMemoryTextFromRequestBody(
-  body: Record<string, unknown> | null | undefined
+  body: Record<string, unknown> | null | undefined,
+  videoTranscriptSensitive = false,
+  context: VideoTranscriptLogContext = {}
 ): string {
   if (!body || typeof body !== "object") return "";
+  // Re-check the structured body at the sink boundary. The explicit bit covers
+  // processed requests whose raw carrier was already replaced; trusted hashes
+  // identify only descriptions emitted by a modified Video Bridge guardrail.
+  const transcriptSensitive =
+    videoTranscriptSensitive || containsVideoTranscriptForLog(body, context);
 
   const messages = Array.isArray(body.messages) ? body.messages : null;
   if (messages && messages.length > 0) {
@@ -39,18 +80,16 @@ export function extractMemoryTextFromRequestBody(
       const msg = messages[i] as Record<string, unknown>;
       if (msg?.role !== "user") continue;
 
-      if (typeof msg.content === "string" && msg.content.trim().length > 0) {
+      const messageText = normalizeMemoryInputText(msg.content, context);
-        return capMemoryExtractionText(msg.content.trim());
+      if (messageText) {
+        return capMemoryExtractionText(messageText);
       }
 
       if (Array.isArray(msg.content)) {
         const text = msg.content
-          .map((part: Record<string, unknown>) => {
+          .map((part: Record<string, unknown>) =>
-            if (typeof part?.text === "string") return part.text.trim();
+            extractMemoryTextPart(part, transcriptSensitive, context)
-            if (part?.type === "input_text" && typeof part?.text === "string")
+          )
-              return part.text.trim();
-            return "";
-          })
           .filter(Boolean)
           .join("\n")
           .trim();
@@ -68,17 +107,15 @@ export function extractMemoryTextFromRequestBody(
       if (role && role !== "user") continue;
       if (itemType && itemType !== "message") continue;
 
-      if (typeof item?.content === "string" && item.content.trim()) {
+      const itemText = normalizeMemoryInputText(item?.content, context);
-        return capMemoryExtractionText(item.content.trim());
+      if (itemText) {
+        return capMemoryExtractionText(itemText);
       }
       if (Array.isArray(item?.content)) {
         const text = item.content
-          .map((part: Record<string, unknown>) => {
+          .map((part: Record<string, unknown>) =>
-            if (typeof part?.text === "string") return part.text.trim();
+            extractMemoryTextPart(part, transcriptSensitive, context)
-            if (part?.type === "input_text" && typeof part?.text === "string")
+          )
-              return part.text.trim();
-            return "";
-          })
           .filter(Boolean)
           .join("\n")
           .trim();
@@ -96,15 +133,14 @@ export function extractMemoryTextFromRequestBody(
         if (role && role !== "user") return "";
         if (itemType && itemType !== "message") return "";
 
-        if (typeof item?.content === "string") return item.content.trim();
+        if (typeof item?.content === "string") {
+          return normalizeMemoryInputText(item.content, context);
+        }
         if (Array.isArray(item?.content)) {
           return item.content
-            .map((part: Record<string, unknown>) => {
+            .map((part: Record<string, unknown>) =>
-              if (typeof part?.text === "string") return part.text.trim();
+              extractMemoryTextPart(part, transcriptSensitive, context)
-              if (part?.type === "input_text" && typeof part?.text === "string")
+            )
-                return part.text.trim();
-              return "";
-            })
             .filter(Boolean)
             .join("\n")
             .trim();

open-sse/handlers/chatCore/pluginOnRequest.ts

@@ -9,6 +9,8 @@
  * byte-identical to the previous inline block.
  */
 
+import { redactVideoTranscriptSensitiveText } from "@/lib/guardrails/videoTranscriptLogRedaction";
+
 type LoggerLike =
   { info?: (...args: unknown[]) => void; debug?: (...args: unknown[]) => void } | null | undefined;
 
@@ -25,6 +27,7 @@ export async function runPluginOnRequestHook(args: {
   apiKeyInfo: unknown;
   headers?: Record<string, string | string[] | undefined>;
   log?: LoggerLike;
+  videoTranscriptSensitive?: boolean;
 }): Promise<PluginOnRequestGate> {
   try {
     const { runOnRequest } = await import("@/lib/plugins/hooks");
@@ -36,6 +39,7 @@ export async function runPluginOnRequestHook(args: {
       apiKeyInfo: args.apiKeyInfo,
       headers: args.headers,
       metadata: {},
+      videoTranscriptSensitive: args.videoTranscriptSensitive === true,
     };
     const pluginResult = await runOnRequest(pluginCtx);
     if (pluginResult?.blocked) {
@@ -55,10 +59,11 @@ export async function runPluginOnRequestHook(args: {
     }
     return { blocked: false, body: pluginResult?.body };
   } catch (pluginErr) {
-    args.log?.debug?.(
+    const retainedPluginError = redactVideoTranscriptSensitiveText(
-      "PLUGIN",
+      pluginErr instanceof Error ? pluginErr.message : String(pluginErr),
-      `onRequest hook error (non-fatal): ${pluginErr instanceof Error ? pluginErr.message : String(pluginErr)}`
+      args.videoTranscriptSensitive === true
     );
+    args.log?.debug?.("PLUGIN", `onRequest hook error (non-fatal): ${retainedPluginError}`);
     return { blocked: false };
   }
 }

open-sse/handlers/chatCore/pluginOnResponse.ts

@@ -26,6 +26,7 @@ export async function runPluginOnResponseHook(args: {
   apiKeyInfo: unknown;
   headers?: Record<string, string | string[] | undefined>;
   response: PluginOnResponsePayload;
+  videoTranscriptSensitive?: boolean;
 }): Promise<void> {
   try {
     const { runOnResponse } = await import("@/lib/plugins/hooks");
@@ -38,6 +39,7 @@ export async function runPluginOnResponseHook(args: {
         apiKeyInfo: args.apiKeyInfo,
         headers: args.headers,
         metadata: {},
+        videoTranscriptSensitive: args.videoTranscriptSensitive === true,
       },
       args.response
     ).catch(() => {});

open-sse/handlers/chatCore/postCallGuardrailContext.ts

@@ -36,6 +36,7 @@ export function buildPostCallGuardrailContext(
     provider: string | null | undefined;
     responsePayloadFormat: unknown;
     clientResponseFormat: unknown;
+    videoTranscriptSensitive?: boolean;
   },
   resolveDisabledGuardrails: typeof defaultResolveDisabled = defaultResolveDisabled
 ): GuardrailContext {
@@ -57,5 +58,6 @@ export function buildPostCallGuardrailContext(
     sourceFormat: optionalString(args.responsePayloadFormat),
     stream: false,
     targetFormat: optionalString(args.clientResponseFormat),
+    videoTranscriptSensitive: args.videoTranscriptSensitive === true,
   };
 }

open-sse/handlers/chatCore/semanticCache.ts

@@ -1,8 +1,4 @@
-import {
+import { generateSignature, getCachedResponse, isCacheableForRead } from "@/lib/semanticCache";
-  generateSignature,
-  getCachedResponse,
-  isCacheableForRead,
-} from "@/lib/semanticCache";
 import { calculateCost } from "@/lib/usage/costCalculator";
 import { trackPendingRequest } from "@/lib/usageDb";
 import { synthesizeOpenAiSseFromJson } from "../../utils/jsonToSse.ts";
@@ -25,6 +21,7 @@ export async function checkSemanticCache({
   persistAttemptLogs,
   apiKeyId,
   cacheDefaultMode,
+  videoTranscriptSensitive,
 }: {
   semanticCacheEnabled: boolean;
   // Only the fields this read path actually touches are named; everything else
@@ -42,7 +39,9 @@ export async function checkSemanticCache({
   persistAttemptLogs: (args: unknown) => void;
   apiKeyId?: string | null;
   cacheDefaultMode?: "legacy" | "bypass" | null;
+  videoTranscriptSensitive: boolean;
 }) {
+  if (videoTranscriptSensitive) return null;
   // Per-key bypass: skip cache lookup entirely when the API key opts out.
   if (cacheDefaultMode === "bypass") return null;
   if (semanticCacheEnabled && isCacheableForRead(body, clientRawRequest?.headers)) {

open-sse/handlers/chatCore/semanticCacheStore.ts

@@ -50,10 +50,12 @@ export function storeSemanticCacheResponse(
     apiKeyId?: string;
     usage?: UsageLike;
     log?: LoggerLike;
+    videoTranscriptSensitive: boolean;
   },
   deps: SemanticCacheStoreDeps = DEFAULT_DEPS
 ): void {
   if (
+    args.videoTranscriptSensitive ||
     !args.enabled ||
     !deps.isCacheableForWrite(args.body, args.headers) ||
     !deps.isSmallEnoughForSemanticCache(args.translatedResponse)

open-sse/handlers/chatCore/streamingPipeline.ts

@@ -68,6 +68,7 @@ export function assembleStreamingPipeline(
     clientRawRequestHeaders: HeadersLike;
     clientResponseFormat: Parameters<typeof defaultShape>[0];
     echoModel: string | null | undefined;
+    redactStreamDiagnosticsForLog?: boolean;
     responseHeaders: Record<string, string>;
   },
   deps: StreamingPipelineDeps = DEFAULT_DEPS
@@ -83,7 +84,8 @@ export function assembleStreamingPipeline(
   let piiStream = deps.pipeWithDisconnect(
     args.providerResponse,
     args.transformStream,
-    args.streamController
+    args.streamController,
+    { redactStreamDiagnosticsForLog: args.redactStreamDiagnosticsForLog }
   );
   if (typeof args.createPiiTransform === "function") {
     piiStream = piiStream.pipeThrough((args.createPiiTransform as () => TransformStream)());

open-sse/handlers/chatCore/streamingSemanticCacheStore.ts

@@ -49,6 +49,7 @@ interface StreamingCacheArgs {
   apiKeyId?: string;
   streamUsage?: Record<string, unknown> | null;
   log?: LoggerLike;
+  videoTranscriptSensitive: boolean;
 }
 
 function streamTokensSaved(streamUsage: Record<string, unknown> | null | undefined): number {
@@ -87,6 +88,7 @@ export function storeStreamingSemanticCacheResponse(
   deps: StreamingSemanticCacheStoreDeps = DEFAULT_DEPS
 ): void {
   if (
+    args.videoTranscriptSensitive ||
     !args.enabled ||
     args.streamStatus !== 200 ||
     !args.streamResponseBody ||

open-sse/services/combo.ts

@@ -63,6 +63,10 @@ import {
 } from "../../src/lib/db/contextHandoffs.ts";
 import { extractSessionAffinityKey } from "@/sse/services/auth";
 import { getHiddenModelsByProvider } from "@/models";
+import {
+  containsVideoTranscriptForLog,
+  redactVideoTranscriptSensitiveText,
+} from "@/lib/guardrails/videoTranscriptLogRedaction";
 import { resolveModelLockoutSettings } from "../../src/lib/resilience/modelLockoutSettings";
 import { fetchCodexQuota } from "./codexQuotaFetcher.ts";
 import { evaluateQuotaCutoff, getQuotaFetcher, type QuotaInfo } from "./quotaPreflight.ts";
@@ -702,7 +706,15 @@ export async function resolveTargetTimeoutMsForTarget(
  */
 export async function handleComboChat(options: HandleComboChatOptions): Promise<Response> {
   const traceInvocationId = options.invocationId ?? createInvocationId();
-  const response = await handleComboChatInner({ ...options, invocationId: traceInvocationId });
+  // Re-check a still-structured body at the service boundary. The explicit bit remains
+  // authoritative when the caller already ran Video Bridge and replaced the raw carrier.
+  const videoTranscriptSensitive =
+    options.videoTranscriptSensitive === true || containsVideoTranscriptForLog(options.body);
+  const response = await handleComboChatInner({
+    ...options,
+    invocationId: traceInvocationId,
+    videoTranscriptSensitive,
+  });
   response.headers.set("X-OmniRoute-Combo-Trace", traceInvocationId);
   const trace = getComboTrace(traceInvocationId);
   options.log.info(
@@ -732,6 +744,8 @@ async function handleComboChatInner({
   sourceFormat = null,
   endpointPath = null,
   requestHeaders = null,
+  videoTranscriptSensitive = false,
+  videoTranscriptDescriptionFingerprints = [],
   invocationId,
 }: HandleComboChatOptions): Promise<Response> {
   const comboCtx = createComboContext({ body, combo, settings, relayOptions, log });
@@ -783,6 +797,7 @@ async function handleComboChatInner({
       handleSingleModelWithTimeout,
       log,
       hiddenModelsByProvider,
+      videoTranscriptSensitive,
     });
     if (pinnedDispatch) return pinnedDispatch;
   }
@@ -805,6 +820,8 @@ async function handleComboChatInner({
     signal,
     apiKeyAllowedConnections,
     hiddenModelsByProvider,
+    videoTranscriptSensitive,
+    videoTranscriptDescriptionFingerprints,
     perTargetAdmission,
     deferContextOverflowWhenCompressible,
     compressionExclusions,
@@ -861,6 +878,8 @@ async function handleComboChatInner({
     signal,
     apiKeyAllowedConnections,
     hiddenModelsByProvider,
+    videoTranscriptSensitive,
+    videoTranscriptDescriptionFingerprints,
     perTargetAdmission,
     deferContextOverflowWhenCompressible,
     compressionExclusions,
@@ -889,6 +908,7 @@ async function handleComboChatInner({
       allCombos,
       signal,
       hiddenModelsByProvider,
+      videoTranscriptSensitive,
       clientManagedResponsesContext,
       deferContextOverflowWhenCompressible,
       compressionExclusions,
@@ -1681,9 +1701,13 @@ async function handleComboChatInner({
             releaseQualityClone(qualityClone, result, quality);
             if (!quality.valid) {
               releaseRejectedQualityResponse(qualityClone, result);
+              const retainedQualityReason = redactVideoTranscriptSensitiveText(
+                quality.reason || "upstream response failed quality validation",
+                videoTranscriptSensitive
+              );
               log.warn(
                 "COMBO",
-                `Model ${modelStr} returned 200 but failed quality check: ${quality.reason}`
+                `Model ${modelStr} returned 200 but failed quality check: ${retainedQualityReason}`
               );
               // #6692: a quality-rejected 200 never marks the connection row
               // unhealthy, so the sticky pin's lazy headroom recheck would never
@@ -1736,7 +1760,7 @@ async function handleComboChatInner({
                 targetIndex: i,
                 provider,
                 model: modelStr,
-                error: `Quality: ${quality.reason}`,
+                error: `Quality: ${retainedQualityReason}`,
                 latencyMs: Date.now() - startTime,
               });
               observeFailure(false, target.executionKey);
@@ -1869,6 +1893,8 @@ async function handleComboChatInner({
                   prevModel,
                   currModel: modelStr,
                   universalConfig: universalHandoffConfig,
+                  videoTranscriptSensitive,
+                  trustedDescriptionFingerprints: videoTranscriptDescriptionFingerprints,
                   handleSingleModel: handleSingleModelWithTimeout,
                 });
               }
@@ -1920,6 +1946,8 @@ async function handleComboChatInner({
                     model: modelStr,
                     expiresAt: resetCandidates[0] || null,
                     config: relayConfig,
+                    videoTranscriptSensitive,
+                    trustedDescriptionFingerprints: videoTranscriptDescriptionFingerprints,
                     handleSingleModel: handleSingleModelWithTimeout,
                   });
                 }
@@ -2443,7 +2471,9 @@ async function handleComboChatInner({
           }
           log.warn("COMBO", `Model ${modelStr} failed, trying next`, {
             status: result.status,
-            errorBody: redactConnectionLabel(errorText),
+            errorBody: redactConnectionLabel(
+              redactVideoTranscriptSensitiveText(errorText, videoTranscriptSensitive)
+            ),
           });
 
           // #5976: per-model-quota providers (Gemini, GitHub, etc.) multiplex models
@@ -2524,7 +2554,11 @@ async function handleComboChatInner({
           }
         })().catch((err) => {
           const logError = log.error ?? log.warn;
-          logError("COMBO", `Speculative task error for target ${i}`, err);
+          logError(
+            "COMBO",
+            `Speculative task error for target ${i}`,
+            videoTranscriptSensitive ? redactVideoTranscriptSensitiveText(String(err), true) : err
+          );
           // G2 (silent-stop fix): never leave the speculative loop waiting on an
           // unresolved globalPromise. If a task throws unexpectedly (outside
           // executeTarget's error handling) and no other task succeeds, the post-loop
@@ -2745,9 +2779,13 @@ async function handleComboChatInner({
         });
 
         if (decision.wait) {
+          const retainedComboFailure = redactVideoTranscriptSensitiveText(
+            msg,
+            videoTranscriptSensitive
+          );
           log.info(
             "COMBO",
-            `${strategy} cooldown wait: ${msg} — waiting ${Math.ceil(
+            `${strategy} cooldown wait: ${retainedComboFailure} — waiting ${Math.ceil(
               decision.waitMs / 1000
             )}s (reason=${decision.reason ?? "?"}) then retrying (attempt ${
               comboCooldownAttempt + 1
@@ -2773,7 +2811,13 @@ async function handleComboChatInner({
       // a config-class status like 403/422).
       if (earliestRetryAfter && isRetryAfterEligibleStatus(status)) {
         const retryHuman = formatRetryAfter(toRetryAfterDisplayValue(earliestRetryAfter));
-        log.warn("COMBO", `All models failed | ${msg} (${retryHuman})`);
+        log.warn(
+          "COMBO",
+          `All models failed | ${redactVideoTranscriptSensitiveText(
+            msg,
+            videoTranscriptSensitive
+          )} (${retryHuman})`
+        );
         return withQuotaExhaustionClassification(
           unavailableResponse(status, msg, earliestRetryAfter, retryHuman),
           observedFailure ? allObservedFailuresQuota : null
@@ -2784,7 +2828,10 @@ async function handleComboChatInner({
       // `try-auto` recovery action via buildRecoveryHint so the OC plugin can show "→ Try
       // model: auto" instead of an opaque 5xx. We pass the upstream retry-after seconds to
       // the hint so the client can render a precise "wait Ns and retry" message.
-      log.warn("COMBO", `All models failed | ${msg}`);
+      log.warn(
+        "COMBO",
+        `All models failed | ${redactVideoTranscriptSensitiveText(msg, videoTranscriptSensitive)}`
+      );
       const { pinClearedNow } = recordComboFailure(effectiveSessionId, combo.name);
       if (pinClearedNow) {
         log.info(
@@ -2888,6 +2935,7 @@ async function handleRoundRobinCombo({
   requestHeaders = null,
   relayOptions,
   perTargetAdmission = null,
+  videoTranscriptSensitive = false,
 }: HandleRoundRobinOptions): Promise<Response> {
   const config = settings
     ? resolveComboConfig(combo, settings)
@@ -3286,24 +3334,20 @@ async function handleRoundRobinCombo({
               "COMBO-RR",
               `Maximum combo attempts (${maxGlobalAttempts}) exceeded. Terminating loop to prevent runaway requests.`
             );
-            return errorResponseWithComboDiagnostics(
+            return errorResponseWithComboDiagnostics(503, "Maximum combo retry limit reached", {
-              503,
+              poolSize: modelCount,
-              "Maximum combo retry limit reached",
+              attempted: globalAttempts,
-              {
+              excluded: [
-                poolSize: modelCount,
+                ...[...exhaustedProviders].map((p) => ({ provider: p, reason: "exhausted" })),
-                attempted: globalAttempts,
+                ...[...exhaustedConnections].map((c) => formatExhaustedConnectionKey(String(c))),
-                excluded: [
+              ],
-                  ...[...exhaustedProviders].map((p) => ({ provider: p, reason: "exhausted" })),
+              attemptOrder: rrOutcomes.map((o) => ({
-                  ...[...exhaustedConnections].map((c) => formatExhaustedConnectionKey(String(c))),
+                provider: o.model.split("/")[0] || "unknown",
-                ],
+                model: o.model,
-                attemptOrder: rrOutcomes.map((o) => ({
+              })),
-                  provider: o.model.split("/")[0] || "unknown",
+              terminalReason: "max_attempts_exceeded",
-                  model: o.model,
+              recovery: buildRecoveryHint("max_attempts_exceeded"),
-                })),
+            });
-                terminalReason: "max_attempts_exceeded",
-                recovery: buildRecoveryHint("max_attempts_exceeded"),
-              }
-            );
           }
           if (retry > 0) {
             log.info(
@@ -3404,9 +3448,13 @@ async function handleRoundRobinCombo({
             releaseQualityClone(rrClone, result, quality);
             if (!quality.valid) {
               releaseRejectedQualityResponse(rrClone, result);
+              const retainedQualityReason = redactVideoTranscriptSensitiveText(
+                quality.reason || "upstream response failed quality validation",
+                videoTranscriptSensitive
+              );
               log.warn(
                 "COMBO-RR",
-                `${modelStr} returned 200 but failed quality check: ${quality.reason}`
+                `${modelStr} returned 200 but failed quality check: ${retainedQualityReason}`
               );
               // #6692: same rationale as handleComboChat's quality-fail branch —
               // a quality-rejected 200 never marks the connection row unhealthy,
@@ -3758,7 +3806,9 @@ async function handleRoundRobinCombo({
           if (offset > 0) fallbackCount++;
           log.warn("COMBO-RR", `${modelStr} failed, trying next model`, {
             status: result.status,
-            errorBody: redactConnectionLabel(errorText),
+            errorBody: redactConnectionLabel(
+              redactVideoTranscriptSensitiveText(errorText, videoTranscriptSensitive)
+            ),
           });
 
           if (
@@ -3811,7 +3861,11 @@ async function handleRoundRobinCombo({
   } catch (err) {
     // G4: unexpected exception in the round-robin loop must never crash the
     // request silently — surface a 500 instead of hanging the client.
-    log.error?.("COMBO-RR", "Unexpected error in round-robin loop", err);
+    log.error?.(
+      "COMBO-RR",
+      "Unexpected error in round-robin loop",
+      videoTranscriptSensitive ? redactVideoTranscriptSensitiveText(String(err), true) : err
+    );
     return errorResponse(500, "Unexpected error in round-robin combo");
   } finally {
     if (rrLoopSafetyTimer) {
@@ -3907,11 +3961,20 @@ async function handleRoundRobinCombo({
 
   if (earliestRetryAfter && isRetryAfterEligibleStatus(status)) {
     const retryHuman = formatRetryAfter(toRetryAfterDisplayValue(earliestRetryAfter));
-    log.warn("COMBO-RR", `All models failed | ${msg} (${retryHuman})`);
+    log.warn(
+      "COMBO-RR",
+      `All models failed | ${redactVideoTranscriptSensitiveText(
+        msg,
+        videoTranscriptSensitive
+      )} (${retryHuman})`
+    );
     return unavailableResponse(status, msg, earliestRetryAfter, retryHuman);
   }
 
-  log.warn("COMBO-RR", `All models failed | ${msg}`);
+  log.warn(
+    "COMBO-RR",
+    `All models failed | ${redactVideoTranscriptSensitiveText(msg, videoTranscriptSensitive)}`
+  );
   return new Response(JSON.stringify({ error: { message: msg } }), {
     status,
     headers: { "Content-Type": "application/json" },

open-sse/services/combo/dispatchPrelude.ts

@@ -13,6 +13,7 @@
  * Extracted from combo.ts as a pure move (#3501). No behaviour change.
  */
 import { getCachedProviderConnections } from "../../../src/lib/db/readCache";
+import { redactVideoTranscriptSensitiveText } from "../../../src/lib/guardrails/videoTranscriptLogRedaction";
 import { getCircuitBreaker } from "../../../src/shared/utils/circuitBreaker";
 import { fisherYatesShuffle, getNextFromDeck } from "../../../src/shared/utils/shuffleDeck";
 import { handleFusionChat, type FusionTuning } from "../fusion.ts";
@@ -84,6 +85,8 @@ type PreludeBaseOptionArgs = {
   signal?: AbortSignal | null;
   apiKeyAllowedConnections?: string[] | null;
   hiddenModelsByProvider?: HiddenModelsByProvider;
+  videoTranscriptSensitive?: boolean;
+  videoTranscriptDescriptionFingerprints?: readonly string[];
   clientManagedResponsesContext?: boolean;
   /** #9654 Wave 2: per-target lane-aware admission probe (see HandleComboChatOptions). */
   perTargetAdmission?: PerTargetAdmissionHook | null;
@@ -111,6 +114,8 @@ function buildBaseOptions(a: PreludeBaseOptionArgs): HandleComboChatOptions {
     signal: a.signal,
     apiKeyAllowedConnections: a.apiKeyAllowedConnections,
     hiddenModelsByProvider: a.hiddenModelsByProvider,
+    videoTranscriptSensitive: a.videoTranscriptSensitive,
+    videoTranscriptDescriptionFingerprints: a.videoTranscriptDescriptionFingerprints,
     invocationId: a.invocationId,
     clientManagedResponsesContext: a.clientManagedResponsesContext,
     perTargetAdmission: a.perTargetAdmission,
@@ -221,8 +226,16 @@ async function evaluatePinnedResponse(args: {
   clientRequestedStream: boolean;
   config: ComboSetupConfig;
   log: ComboLogger;
+  videoTranscriptSensitive?: boolean;
 }): Promise<Response | null> {
-  const { pinnedResult, pinnedModel, clientRequestedStream, config, log } = args;
+  const {
+    pinnedResult,
+    pinnedModel,
+    clientRequestedStream,
+    config,
+    log,
+    videoTranscriptSensitive = false,
+  } = args;
   if (pinnedResult.ok) {
     let pinnedClone: Response;
     try {
@@ -239,9 +252,13 @@ async function evaluatePinnedResponse(args: {
     releaseQualityClone(pinnedClone, pinnedResult, pinnedQuality);
     if (pinnedQuality.valid) return pinnedResult;
     releaseRejectedQualityResponse(pinnedClone, pinnedResult);
+    const retainedReason = redactVideoTranscriptSensitiveText(
+      pinnedQuality.reason || "upstream response failed quality validation",
+      videoTranscriptSensitive
+    );
     log.warn(
       "COMBO",
-      `Pinned model ${pinnedModel} returned 200 but failed quality check: ${pinnedQuality.reason}, falling through to combo retry/fallback`
+      `Pinned model ${pinnedModel} returned 200 but failed quality check: ${retainedReason}, falling through to combo retry/fallback`
     );
     return null;
   }
@@ -271,6 +288,7 @@ export async function tryPinnedModelDispatch(args: {
   handleSingleModelWithTimeout: HandleSingleModel;
   log: ComboLogger;
   hiddenModelsByProvider?: HiddenModelsByProvider;
+  videoTranscriptSensitive?: boolean;
 }): Promise<Response | null> {
   const {
     body,
@@ -282,6 +300,7 @@ export async function tryPinnedModelDispatch(args: {
     handleSingleModelWithTimeout,
     log,
     hiddenModelsByProvider,
+    videoTranscriptSensitive = false,
   } = args;
   // The pin is read from session_model_history (a PRIOR turn) and may name a
   // model that has since been removed from this combo, or a provider whose
@@ -341,9 +360,13 @@ export async function tryPinnedModelDispatch(args: {
         modelPinned: true,
       } as SingleModelTarget);
     } catch (pinErr) {
+      const retainedPinError = redactVideoTranscriptSensitiveText(
+        pinErr instanceof Error ? pinErr.message : String(pinErr),
+        videoTranscriptSensitive
+      );
       log.warn(
         "COMBO",
-        `Pinned model ${pinnedModel} threw error: ${pinErr instanceof Error ? pinErr.message : String(pinErr)}, falling through to combo retry/fallback`
+        `Pinned model ${pinnedModel} threw error: ${retainedPinError}, falling through to combo retry/fallback`
       );
     }
     if (pinnedResult) {
@@ -353,6 +376,7 @@ export async function tryPinnedModelDispatch(args: {
         clientRequestedStream,
         config,
         log,
+        videoTranscriptSensitive,
       });
       if (accepted) return accepted;
     }
@@ -395,6 +419,8 @@ export async function tryFusionDispatch(args: {
   signal?: AbortSignal | null;
   apiKeyAllowedConnections?: string[] | null;
   hiddenModelsByProvider?: HiddenModelsByProvider;
+  videoTranscriptSensitive?: boolean;
+  videoTranscriptDescriptionFingerprints?: readonly string[];
   perTargetAdmission?: PerTargetAdmissionHook | null;
   deferContextOverflowWhenCompressible?: boolean;
   compressionExclusions?: import("../compression/exclusions.ts").CompressionExclusions;
@@ -524,6 +550,7 @@ export async function tryFusionDispatch(args: {
     perTargetAdmission: args.perTargetAdmission,
     judgeModel,
     tuning: fusionTuning,
+    videoTranscriptSensitive: args.videoTranscriptSensitive,
   });
 }
 
@@ -676,6 +703,8 @@ export async function tryRuntimeUnitDispatch(args: {
   signal?: AbortSignal | null;
   apiKeyAllowedConnections?: string[] | null;
   hiddenModelsByProvider?: HiddenModelsByProvider;
+  videoTranscriptSensitive?: boolean;
+  videoTranscriptDescriptionFingerprints?: readonly string[];
   perTargetAdmission?: PerTargetAdmissionHook | null;
   deferContextOverflowWhenCompressible?: boolean;
   compressionExclusions?: import("../compression/exclusions.ts").CompressionExclusions;

open-sse/services/combo/types.ts

@@ -114,6 +114,10 @@ export type HandleComboChatOptions = {
   apiKeyAllowedConnections?: string[] | null;
   nesting?: ComboNestingContext | null;
   hiddenModelsByProvider?: HiddenModelsByProvider;
+  /** Request-scoped retention bit derived before guardrails can replace a video carrier. */
+  videoTranscriptSensitive?: boolean;
+  /** Exact bounded identities emitted by a modified Video Bridge guardrail. */
+  videoTranscriptDescriptionFingerprints?: readonly string[];
   /** Native Responses clients (for example Codex CLI/Desktop) manage compaction themselves. */
   clientManagedResponsesContext?: boolean;
   /**

open-sse/services/contextHandoff.ts

@@ -5,6 +5,11 @@ import {
   type HandoffPayload,
   upsertHandoff,
 } from "../../src/lib/db/contextHandoffs.ts";
+import {
+  containsVideoTranscriptForLog,
+  omitVideoTranscriptForLog,
+  type VideoTranscriptLogContext,
+} from "../../src/lib/guardrails/videoTranscriptLogRedaction.ts";
 import { estimateTokens } from "./contextManager.ts";
 import { stripMarkdownCodeFence } from "../utils/aiSdkCompat.ts";
 
@@ -43,6 +48,23 @@ export type MessageLike = {
   content?: unknown;
 };
 
+type HandoffTranscriptContext = VideoTranscriptLogContext & {
+  videoTranscriptSensitive?: boolean;
+};
+
+function retainHandoffMessages(
+  messages: MessageLike[],
+  context: HandoffTranscriptContext
+): MessageLike[] | null {
+  const source = Array.isArray(messages) ? messages : [];
+  const transcriptSensitive =
+    context.videoTranscriptSensitive === true || containsVideoTranscriptForLog(source, context);
+  if (!transcriptSensitive) return source;
+
+  const retained = omitVideoTranscriptForLog(source, context);
+  return Array.isArray(retained) ? (retained as MessageLike[]) : null;
+}
+
 export interface ContextRelayConfig {
   handoffModel?: string;
   handoffThreshold?: number;
@@ -445,6 +467,10 @@ export function maybeGenerateHandoff(options: {
   model: string;
   expiresAt: string | null;
   config?: ContextRelayConfig | null;
+  /** Trusted request bit for carriers already replaced by the Video Bridge guardrail. */
+  videoTranscriptSensitive?: boolean;
+  /** Exact bounded identities emitted by a modified Video Bridge guardrail. */
+  trustedDescriptionFingerprints?: readonly string[];
   handleSingleModel: (body: Record<string, unknown>, modelStr: string) => Promise<Response>;
 }): void {
   if (!options.sessionId || !options.connectionId) return;
@@ -454,6 +480,12 @@ export function maybeGenerateHandoff(options: {
   if (options.percentUsed < relayConfig.handoffThreshold) return;
   if (options.percentUsed >= HANDOFF_EXHAUSTION_THRESHOLD) return;
 
+  const retainedMessages = retainHandoffMessages(options.messages, {
+    videoTranscriptSensitive: options.videoTranscriptSensitive,
+    trustedDescriptionFingerprints: options.trustedDescriptionFingerprints,
+  });
+  if (!retainedMessages) return;
+
   cleanupExpiredHandoffs();
   if (hasActiveHandoff(options.sessionId, options.comboName)) return;
   const inflightKey = getInflightKey(options.sessionId, options.comboName);
@@ -463,6 +495,7 @@ export function maybeGenerateHandoff(options: {
   setImmediate(() => {
     generateHandoffAsync({
       ...options,
+      messages: retainedMessages,
       sessionId: options.sessionId as string,
       connectionId: options.connectionId as string,
       config: relayConfig,
@@ -690,6 +723,10 @@ export function maybeGenerateUniversalHandoff(options: {
   prevModel: string | null;
   currModel: string;
   universalConfig: UniversalHandoffConfig;
+  /** Trusted request bit for carriers already replaced by the Video Bridge guardrail. */
+  videoTranscriptSensitive?: boolean;
+  /** Exact bounded identities emitted by a modified Video Bridge guardrail. */
+  trustedDescriptionFingerprints?: readonly string[];
   handleSingleModel: (body: Record<string, unknown>, modelStr: string) => Promise<Response>;
 }): void {
   const decision = shouldGenerateUniversalHandoff({
@@ -703,6 +740,12 @@ export function maybeGenerateUniversalHandoff(options: {
   if (decision !== "generate") return;
   if (!options.sessionId) return;
 
+  const retainedMessages = retainHandoffMessages(options.messages, {
+    videoTranscriptSensitive: options.videoTranscriptSensitive,
+    trustedDescriptionFingerprints: options.trustedDescriptionFingerprints,
+  });
+  if (!retainedMessages) return;
+
   const inflightKey = getInflightKey(options.sessionId, options.comboName);
   if (inflightHandoffGenerations.has(inflightKey)) return;
   inflightHandoffGenerations.add(inflightKey);
@@ -713,7 +756,7 @@ export function maybeGenerateUniversalHandoff(options: {
     generateUniversalHandoffAsync({
       sessionId: options.sessionId as string,
       comboName: options.comboName,
-      messages: options.messages,
+      messages: retainedMessages,
       prevModel: options.prevModel || "unknown",
       currModel: options.currModel,
       handoffModel: options.universalConfig.handoffModel || options.currModel,

open-sse/services/fusion.ts

@@ -20,6 +20,7 @@
  */
 import { errorResponse, sanitizeErrorMessage } from "../utils/error.ts";
 import { extractTextContent } from "../translator/helpers/geminiHelper.ts";
+import { redactVideoTranscriptSensitiveText } from "../../src/lib/guardrails/videoTranscriptLogRedaction.ts";
 import type { PerTargetAdmissionHook } from "./admission/types.ts";
 import type { ComboLogger, HandleSingleModel, ResolvedComboTarget } from "./combo/types.ts";
 
@@ -231,6 +232,8 @@ export type HandleFusionChatOptions = {
   tuning?: FusionTuning | null;
   /** #9654 Wave 2: per-target lane-aware admission probe (see HandleComboChatOptions). */
   perTargetAdmission?: PerTargetAdmissionHook | null;
+  /** Request-scoped bit set before Video Bridge can replace the raw carrier. */
+  videoTranscriptSensitive?: boolean;
 };
 
 function getFusionModelString(model: FusionModel): string {
@@ -276,6 +279,7 @@ export async function handleFusionChat({
   judgeTarget,
   tuning,
   perTargetAdmission,
+  videoTranscriptSensitive = false,
 }: HandleFusionChatOptions): Promise<Response> {
   const panel = Array.isArray(models) ? models.filter(Boolean) : [];
   if (panel.length === 0) {
@@ -400,7 +404,10 @@ export async function handleFusionChat({
     }
     if (sentinel.__error) {
       log.warn("FUSION", `Panel ${model} threw`, {
-        error: sanitizeErrorMessage(sentinel.__error as Error),
+        error: redactVideoTranscriptSensitiveText(
+          sanitizeErrorMessage(sentinel.__error as Error),
+          videoTranscriptSensitive
+        ),
       });
       failures.push({ model, reason: "threw" });
       continue;
@@ -428,7 +435,10 @@ export async function handleFusionChat({
       }
     } catch (e) {
       log.warn("FUSION", `Panel ${model} unparseable`, {
-        error: sanitizeErrorMessage(e as Error),
+        error: redactVideoTranscriptSensitiveText(
+          sanitizeErrorMessage(e as Error),
+          videoTranscriptSensitive
+        ),
       });
       failures.push({ model, reason: "unparseable" });
     }

open-sse/services/reasoningCache.ts

@@ -140,6 +140,7 @@ type AssistantMessageLike = {
 type AssistantMessageCacheContext = {
   scope?: string;
   historyMessages?: AssistantMessageLike[];
+  videoTranscriptSensitive?: boolean;
 };
 
 type ToolCallLike = {
@@ -334,6 +335,7 @@ export function cacheReasoningFromAssistantMessage(
   model: string,
   context?: AssistantMessageCacheContext
 ): number {
+  if (context?.videoTranscriptSensitive === true) return 0;
   if (!message || message.role !== "assistant") {
     return 0;
   }

open-sse/utils/requestLogger.ts

@@ -1,5 +1,11 @@
 import { getPendingById } from "@/lib/usage/usageHistory";
 import { getChatLogMaxDepth, getChatLogArrayTailItems } from "@/lib/logEnv";
+import {
+  containsVideoTranscriptForLog,
+  omitVideoTranscriptForLog,
+  type VideoTranscriptLogContext,
+  VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+} from "@/lib/guardrails/videoTranscriptLogRedaction";
 import { sanitizeErrorMessage } from "./error.ts";
 
 type JsonRecord = Record<string, unknown>;
@@ -12,6 +18,7 @@ type HeaderInput =
   | undefined;
 
 export type RequestPipelinePayloads = {
+  _omnirouteVideoTranscriptRedacted?: true;
   routeDecision?: JsonRecord;
   clientRawRequest?: JsonRecord;
   openaiRequest?: JsonRecord;
@@ -44,6 +51,7 @@ type RequestLogger = {
   appendConvertedChunk: (chunk: string) => void;
   logError: (error: unknown, requestBody?: unknown) => void;
   getPipelinePayloads: () => RequestPipelinePayloads | null;
+  isVideoTranscriptSensitive: () => boolean;
 };
 
 type RequestLoggerOptions = {
@@ -55,6 +63,13 @@ type RequestLoggerOptions = {
   model?: string;
   provider?: string;
   connectionId?: string | null;
+  /**
+   * Server-derived from a recognized video carrier, successful bridge result, or bounded
+   * unknown-sensitive detector overflow; never from delimiter-shaped prose alone.
+   */
+  videoTranscriptSensitive?: boolean;
+  /** Exact SHA-256 identities of transcript descriptions generated by the guardrail. */
+  videoTranscriptDescriptionFingerprints?: readonly string[];
 };
 
 const DEFAULT_MAX_STREAM_CHUNK_BYTES = 128 * 1024;
@@ -159,7 +174,7 @@ function truncateLogString(value: string, maxLength = MAX_LOG_STRING_LENGTH): st
  * recursing into an object's values, enabling the per-field exemption above.
  * Top-level arrays (no key context) remain subject to truncation.
  */
-export function cloneBoundedForLog(value: unknown, depth = 0, key: string | null = null): unknown {
+function cloneBoundedForLogValue(value: unknown, depth = 0, key: string | null = null): unknown {
   if (value === null || value === undefined) return value;
   if (typeof value === "string") return truncateLogString(value);
   if (typeof value !== "object") return value;
@@ -178,12 +193,15 @@ export function cloneBoundedForLog(value: unknown, depth = 0, key: string | null
     // item and rewrite originalLength with the truncated length (25 instead of the true 800), so
     // the log would misreport how much was cut. Keep the original marker, re-bound only the tail.
     if (isTruncatedArrayMarker(value[0])) {
-      return [value[0], ...value.slice(1).map((item) => cloneBoundedForLog(item, depth + 1))];
+      return [
+        value[0],
+        ...value.slice(1).map((item) => cloneBoundedForLogValue(item, depth + 1, null)),
+      ];
     }
     const exempt = key === "tools";
     const shouldTruncate = !exempt && value.length > MAX_LOG_ARRAY_ITEMS;
     const source = shouldTruncate ? value.slice(-MAX_LOG_ARRAY_ITEMS) : value;
-    const mapped = source.map((item) => cloneBoundedForLog(item, depth + 1));
+    const mapped = source.map((item) => cloneBoundedForLogValue(item, depth + 1, null));
     if (shouldTruncate) {
       return [
         {
@@ -206,7 +224,7 @@ export function cloneBoundedForLog(value: unknown, depth = 0, key: string | null
     ([k]) => !(carried > 0 && k === TRUNCATED_KEYS_MARKER)
   );
   for (const [k, item] of entries.slice(0, MAX_LOG_OBJECT_KEYS)) {
-    result[k] = cloneBoundedForLog(item, depth + 1, k);
+    result[k] = cloneBoundedForLogValue(item, depth + 1, k);
   }
   const dropped = Math.max(0, entries.length - MAX_LOG_OBJECT_KEYS) + carried;
   if (dropped > 0) {
@@ -215,6 +233,17 @@ export function cloneBoundedForLog(value: unknown, depth = 0, key: string | null
   return result;
 }
 
+export function cloneBoundedForLog(
+  value: unknown,
+  depth = 0,
+  key: string | null = null,
+  descriptionContext: VideoTranscriptLogContext = {}
+): unknown {
+  const transcriptSafeValue =
+    depth === 0 ? omitVideoTranscriptForLog(value, descriptionContext) : value;
+  return cloneBoundedForLogValue(transcriptSafeValue, depth, key);
+}
+
 function appendBoundedChunk(
   chunks: string[],
   bytes: { value: number; truncated: boolean },
@@ -277,7 +306,7 @@ function compactPipelinePayloads(
       continue;
     }
 
-    result[key as keyof RequestPipelinePayloads] = value;
+    (result as Record<string, unknown>)[key] = value;
   }
 
   return hasOwnValues(result) ? result : null;
@@ -298,6 +327,7 @@ function makeStreamChunkMethods(options: RequestLoggerOptions, captureChunks: bo
       ? Number(options.maxStreamChunkItems)
       : DEFAULT_MAX_STREAM_CHUNK_ITEMS;
   let pendingPushed = false;
+  let videoTranscriptSensitive = options.videoTranscriptSensitive === true;
 
   const push = () => {
     if (pendingPushed) return;
@@ -330,7 +360,7 @@ function makeStreamChunkMethods(options: RequestLoggerOptions, captureChunks: bo
   };
 
   const append = (arr: string[], bytes: { value: number; truncated: boolean }, chunk: string) => {
-    if (!captureChunks) return;
+    if (!captureChunks || videoTranscriptSensitive) return;
     push();
     const ts = new Date().toISOString().slice(11, 23);
     appendBoundedChunk(arr, bytes, `[${ts}] ${chunk}`, maxBytes, maxItems);
@@ -348,6 +378,17 @@ function makeStreamChunkMethods(options: RequestLoggerOptions, captureChunks: bo
     appendConvertedChunk(chunk: string) {
       append(streamChunks.client, streamChunkBytes.client, chunk);
     },
+    suppressVideoTranscriptChunks() {
+      videoTranscriptSensitive = true;
+      for (const chunks of Object.values(streamChunks)) chunks.splice(0, chunks.length);
+      for (const state of Object.values(streamChunkBytes)) {
+        state.value = 0;
+        state.truncated = false;
+      }
+    },
+    isVideoTranscriptSensitive() {
+      return videoTranscriptSensitive;
+    },
   };
 }
 
@@ -362,26 +403,58 @@ export async function createRequestLogger(
   // so that active requests always have real-time stream data available via
   // the /api/logs/active endpoint.
   const chunkMethods = makeStreamChunkMethods(options, captureStreamChunks);
+  const descriptionLogContext: VideoTranscriptLogContext = {
+    trustedDescriptionFingerprints: options.videoTranscriptDescriptionFingerprints ?? [],
+  };
+  const suppressTranscriptChunksIfNeeded = (
+    descriptionContext: VideoTranscriptLogContext,
+    ...values: unknown[]
+  ): void => {
+    if (values.some((value) => containsVideoTranscriptForLog(value, descriptionContext))) {
+      chunkMethods.suppressVideoTranscriptChunks();
+    }
+  };
+  const cloneRequestBody = (
+    value: unknown,
+    descriptionContext: VideoTranscriptLogContext
+  ): unknown => cloneBoundedForLog(value, 0, null, descriptionContext);
+  const cloneResponseBody = (value: unknown): unknown =>
+    chunkMethods.isVideoTranscriptSensitive()
+      ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER
+      : cloneBoundedForLog(value);
 
   if (options.enabled === false) {
     let routeDecision: JsonRecord | null = null;
     return {
       sessionPath: null,
-      logClientRawRequest() {},
+      logClientRawRequest(_endpoint, body) {
+        suppressTranscriptChunksIfNeeded({}, body);
+      },
       logRouteDecision(decision) {
         routeDecision = cloneBoundedForLog(decision) as JsonRecord;
       },
-      logOpenAIRequest() {},
+      logOpenAIRequest(body) {
-      logTargetRequest() {},
+        suppressTranscriptChunksIfNeeded(descriptionLogContext, body);
-      logProviderResponse() {},
+      },
+      logTargetRequest(_url, _headers, body) {
+        suppressTranscriptChunksIfNeeded(descriptionLogContext, body);
+      },
+      logProviderResponse(_status, _statusText, _headers, body) {
+        void body;
+      },
       appendProviderChunk: chunkMethods.appendProviderChunk,
       appendOpenAIChunk: chunkMethods.appendOpenAIChunk,
-      logConvertedResponse() {},
+      logConvertedResponse(body) {
+        void body;
+      },
       appendConvertedChunk: chunkMethods.appendConvertedChunk,
-      logError() {},
+      logError(_error, requestBody) {
+        suppressTranscriptChunksIfNeeded(descriptionLogContext, requestBody);
+      },
       getPipelinePayloads() {
         return routeDecision ? { routeDecision } : null;
       },
+      isVideoTranscriptSensitive: chunkMethods.isVideoTranscriptSensitive,
     };
   }
 
@@ -393,11 +466,12 @@ export async function createRequestLogger(
     sessionPath: null,
 
     logClientRawRequest(endpoint, body, headers = {}) {
+      suppressTranscriptChunksIfNeeded({}, body);
       payloads.clientRawRequest = {
         timestamp: new Date().toISOString(),
         endpoint,
         headers: maskSensitiveHeaders(headers),
-        body: cloneBoundedForLog(body),
+        body: cloneRequestBody(body, {}),
       };
     },
 
@@ -406,18 +480,20 @@ export async function createRequestLogger(
     },
 
     logOpenAIRequest(body) {
+      suppressTranscriptChunksIfNeeded(descriptionLogContext, body);
       payloads.openaiRequest = {
         timestamp: new Date().toISOString(),
-        body: cloneBoundedForLog(body),
+        body: cloneRequestBody(body, descriptionLogContext),
       };
     },
 
     logTargetRequest(url, headers, body) {
+      suppressTranscriptChunksIfNeeded(descriptionLogContext, body);
       payloads.providerRequest = {
         timestamp: new Date().toISOString(),
         url,
         headers: maskSensitiveHeaders(headers),
-        body: cloneBoundedForLog(body),
+        body: cloneRequestBody(body, descriptionLogContext),
       };
     },
 
@@ -427,7 +503,7 @@ export async function createRequestLogger(
         status,
         statusText,
         headers: maskSensitiveHeaders(headers),
-        body: cloneBoundedForLog(body),
+        body: cloneResponseBody(body),
       };
     },
 
@@ -436,21 +512,25 @@ export async function createRequestLogger(
     logConvertedResponse(body) {
       payloads.clientResponse = {
         timestamp: new Date().toISOString(),
-        body: cloneBoundedForLog(body),
+        body: cloneResponseBody(body),
       };
     },
     appendConvertedChunk: chunkMethods.appendConvertedChunk,
 
     logError(error, requestBody = null) {
+      suppressTranscriptChunksIfNeeded(descriptionLogContext, requestBody);
       payloads.error = {
         timestamp: new Date().toISOString(),
-        error: sanitizeErrorMessage(error instanceof Error ? error.message : String(error)),
+        error: chunkMethods.isVideoTranscriptSensitive()
-        requestBody: cloneBoundedForLog(requestBody),
+          ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER
+          : sanitizeErrorMessage(error instanceof Error ? error.message : String(error)),
+        requestBody: cloneRequestBody(requestBody, descriptionLogContext),
       };
     },
 
     getPipelinePayloads() {
       return compactPipelinePayloads(payloads);
     },
+    isVideoTranscriptSensitive: chunkMethods.isVideoTranscriptSensitive,
   };
 }

open-sse/utils/stream.ts

@@ -1,6 +1,7 @@
+import { VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER } from "@/lib/guardrails/videoTranscriptLogRedaction";
+import { trackPendingRequest, appendRequestLog } from "@/lib/usageDb";
 import { translateResponse, initState } from "../translator/index.ts";
 import { FORMATS } from "../translator/formats.ts";
-import { trackPendingRequest, appendRequestLog } from "@/lib/usageDb";
 import {
   extractUsage,
   hasValidUsage,
@@ -177,6 +178,8 @@ type StreamOptions = {
    * codex-compatible `namespace` + `name` fields.
    */
   requestToolIdentityMap?: Map<string, { namespace: string; name: string }> | null;
+  /** Omit request-sensitive transcript text from retained stream diagnostics only. */
+  redactStreamDiagnosticsForLog?: boolean;
 };
 
 type TranslateState = ReturnType<typeof initState> & {
@@ -654,7 +657,10 @@ export function createSSEStream(options: StreamOptions = {}) {
     dropResponsesCommentary,
     customToolNames = new Set<string>(),
     requestToolIdentityMap = null,
+    redactStreamDiagnosticsForLog = false,
   } = options;
+  const retainDiagnosticForLog = (value: unknown): unknown =>
+    redactStreamDiagnosticsForLog ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER : value;
   const signatureNamespace = connectionId;
   // Request-body-size metric (for monitoring payload size distribution & correlation with TTFT).
   // The size is JSON-serialised byte count; stored as a performance mark detail so monitoring
@@ -1005,7 +1011,10 @@ export function createSSEStream(options: StreamOptions = {}) {
       try {
         failureHandled = onFailure({ status: 502, message: msg, code: "empty_response" }) === true;
       } catch (e) {
-        console.debug(`[STREAM] onFailure callback error (empty_response):`, e);
+        console.debug(
+          `[STREAM] onFailure callback error (empty_response):`,
+          retainDiagnosticForLog(e)
+        );
       }
     }
     if (decrementPendingRequest && !failureHandled) {
@@ -1199,7 +1208,10 @@ export function createSSEStream(options: StreamOptions = {}) {
                       type: "timeout_error",
                     }) === true;
                 } catch (e) {
-                  console.debug(`[STREAM] onFailure callback error (idle_timeout):`, e);
+                  console.debug(
+                    `[STREAM] onFailure callback error (idle_timeout):`,
+                    retainDiagnosticForLog(e)
+                  );
                 }
               }
               if (!failureHandled) {
@@ -1641,10 +1653,7 @@ export function createSSEStream(options: StreamOptions = {}) {
                         isResponsesCommentaryMessageItem
                       ).items
                     : passthroughResponsesOutputItems;
-                  const backfilled = backfillResponsesCompletedOutput(
+                  const backfilled = backfillResponsesCompletedOutput(parsed, backfillCandidates);
-                    parsed,
-                    backfillCandidates
-                  );
                   const usageNormalized = normalizeUsage(parsed);
                   if (
                     stripped ||
@@ -2040,7 +2049,7 @@ export function createSSEStream(options: StreamOptions = {}) {
                 try {
                   failureHandled = onFailure(failurePayload) === true;
                 } catch (e) {
-                  console.debug(`[STREAM] onFailure callback error:`, e);
+                  console.debug(`[STREAM] onFailure callback error:`, retainDiagnosticForLog(e));
                 }
               }
               clearIdleTimer();
@@ -2624,7 +2633,10 @@ export function createSSEStream(options: StreamOptions = {}) {
                   ),
                 });
               } catch (e) {
-                console.debug(`[STREAM] onComplete callback error (${model || "unknown"}):`, e);
+                console.debug(
+                  `[STREAM] onComplete callback error (${model || "unknown"}):`,
+                  retainDiagnosticForLog(e)
+                );
               }
             } else {
               clearPendingRequestFromStream();
@@ -2712,7 +2724,10 @@ export function createSSEStream(options: StreamOptions = {}) {
                     type: err.type,
                   }) === true;
               } catch (e) {
-                console.debug(`[STREAM] onFailure callback error (${model || "unknown"}):`, e);
+                console.debug(
+                  `[STREAM] onFailure callback error (${model || "unknown"}):`,
+                  retainDiagnosticForLog(e)
+                );
               }
             }
 
@@ -2740,7 +2755,7 @@ export function createSSEStream(options: StreamOptions = {}) {
               } catch (e) {
                 console.debug(
                   `[STREAM] onComplete callback error in error path (${model || "unknown"}):`,
-                  e
+                  retainDiagnosticForLog(e)
                 );
               }
             }
@@ -2945,14 +2960,18 @@ export function createSSEStream(options: StreamOptions = {}) {
             } catch (e) {
               console.debug(
                 `[STREAM] onComplete callback error in flush (${model || "unknown"}):`,
-                e
+                retainDiagnosticForLog(e)
               );
             }
           } else {
             clearPendingRequestFromStream();
           }
         } catch (error) {
-          console.log(`[STREAM] Error in flush (${model || "unknown"}):`, error.message || error);
+          const diagnostic = error instanceof Error ? error.message : error;
+          console.log(
+            `[STREAM] Error in flush (${model || "unknown"}):`,
+            retainDiagnosticForLog(diagnostic)
+          );
         }
       },
       cancel(reason) {
@@ -2982,7 +3001,8 @@ export function createSSETransformStreamWithLogger(
   copilotCompatibleReasoning = false,
   suppressThinkClose = false,
   customToolNames: ReadonlySet<string> = new Set(),
-  requestToolIdentityMap: Map<string, { namespace: string; name: string }> | null = null
+  requestToolIdentityMap: Map<string, { namespace: string; name: string }> | null = null,
+  redactStreamDiagnosticsForLog = false
 ) {
   return createSSEStream({
     mode: STREAM_MODE.TRANSLATE,
@@ -3001,6 +3021,7 @@ export function createSSETransformStreamWithLogger(
     suppressThinkClose,
     customToolNames,
     requestToolIdentityMap,
+    redactStreamDiagnosticsForLog,
   });
 }
 
@@ -3015,7 +3036,8 @@ export function createPassthroughStreamWithLogger(
   apiKeyInfo: unknown = null,
   onFailure: ((payload: StreamFailurePayload) => boolean | void | Promise<void>) | null = null,
   clientResponseFormat: string | null = null,
-  requestToolIdentityMap: Map<string, { namespace: string; name: string }> | null = null
+  requestToolIdentityMap: Map<string, { namespace: string; name: string }> | null = null,
+  redactStreamDiagnosticsForLog = false
 ) {
   return createSSEStream({
     mode: STREAM_MODE.PASSTHROUGH,
@@ -3030,6 +3052,7 @@ export function createPassthroughStreamWithLogger(
     onFailure,
     clientResponseFormat,
     requestToolIdentityMap,
+    redactStreamDiagnosticsForLog,
   });
 }
 

open-sse/utils/streamFailureFinalization.ts

@@ -92,6 +92,7 @@ export function finalizeStreamRequestLog({
   status,
   error,
   errorCode,
+  videoTranscriptSensitive,
   onWarn,
 }: {
   pendingRequestId: string;
@@ -103,6 +104,7 @@ export function finalizeStreamRequestLog({
   status: number;
   error?: string | null;
   errorCode?: string | null;
+  videoTranscriptSensitive: boolean;
   onWarn?: (error: unknown) => void;
 }) {
   try {
@@ -112,6 +114,7 @@ export function finalizeStreamRequestLog({
       status,
       error: error || null,
       errorCode: errorCode || null,
+      videoTranscriptSensitive,
     });
     if (!completedById) {
       finalizeMostRecentPendingRequest(model, provider, connectionId, {
@@ -120,6 +123,7 @@ export function finalizeStreamRequestLog({
         status,
         error: error || null,
         errorCode: errorCode || null,
+        videoTranscriptSensitive,
       });
     }
   } catch (error) {

open-sse/utils/streamHandler.ts

@@ -1,3 +1,4 @@
+import { VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER } from "@/lib/guardrails/videoTranscriptLogRedaction";
 import { trackPendingRequest } from "@/lib/usageDb";
 import { STREAM_IDLE_TIMEOUT_MS } from "../config/constants.ts";
 import { FORMATS } from "../translator/formats.ts";
@@ -39,6 +40,7 @@ type StreamControllerOptions = {
   clientAbortSignal?: AbortSignal | null;
   allowCompletedToolHandoffGrace?: boolean;
   clientDisconnectGracePeriodMs?: number;
+  redactStreamDiagnosticsForLog?: boolean;
 };
 
 type StreamController = ReturnType<typeof createStreamController>;
@@ -243,6 +245,7 @@ export function createStreamController({
   clientAbortSignal,
   allowCompletedToolHandoffGrace = false,
   clientDisconnectGracePeriodMs = 0,
+  redactStreamDiagnosticsForLog = false,
 }: StreamControllerOptions = {}) {
   const abortController = new AbortController();
   const startTime = Date.now();
@@ -253,6 +256,9 @@ export function createStreamController({
   let pendingRequestCleared = false;
   let cleanupClientAbortSignal: (() => void) | null = null;
 
+  const retainDiagnosticForLog = (value: unknown): unknown =>
+    redactStreamDiagnosticsForLog ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER : value;
+
   const logStream = (status) => {
     const duration = Date.now() - startTime;
     const p = provider?.toUpperCase() || "UNKNOWN";
@@ -279,7 +285,7 @@ export function createStreamController({
     } catch (e) {
       console.error(
         `[${getTimeString()}] [streamHandler] trackPendingRequest decrement failed — counter may drift`,
-        e
+        retainDiagnosticForLog(e)
       );
     }
   };
@@ -317,7 +323,7 @@ export function createStreamController({
       disconnected = true;
       cleanupClientAbortListener();
 
-      logStream(`disconnect: ${reason}`);
+      logStream(`disconnect: ${String(retainDiagnosticForLog(reason))}`);
 
       // Decrement pending request counter — the TransformStream flush() won't
       // fire when the client aborts mid-stream, so we must clean up here.
@@ -390,7 +396,7 @@ export function createStreamController({
               duration: Date.now() - startTime,
             }) === true;
         } catch (e) {
-          console.debug(`[STREAM-HANDLER] onError callback error:`, e);
+          console.debug(`[STREAM-HANDLER] onError callback error:`, retainDiagnosticForLog(e));
         }
       }
 
@@ -406,7 +412,7 @@ export function createStreamController({
       }
 
       if (error instanceof Error) {
-        logStream(`error: ${error.message}`);
+        logStream(`error: ${String(retainDiagnosticForLog(error.message))}`);
         return;
       }
       logStream("error: unknown");
@@ -845,9 +851,11 @@ export function pipeWithDisconnect(
   providerResponse: Response,
   transformStream: TransformStream<Uint8Array, Uint8Array>,
   streamController: StreamController,
-  opts: { stallTimeoutMs?: number } = {}
+  opts: { redactStreamDiagnosticsForLog?: boolean; stallTimeoutMs?: number } = {}
 ) {
   const stallTimeoutMs = opts.stallTimeoutMs ?? DEFAULT_STREAM_STALL_TIMEOUT_MS;
+  const retainDiagnosticForLog = (value: unknown): unknown =>
+    opts.redactStreamDiagnosticsForLog ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER : value;
 
   // Watchdog disabled — preserve legacy behavior verbatim.
   if (!stallTimeoutMs || stallTimeoutMs <= 0) {
@@ -887,7 +895,10 @@ export function pipeWithDisconnect(
       try {
         streamController.handleError?.(stallError);
       } catch (e) {
-        console.debug(`[STREAM-HANDLER] stall watchdog handleError failed:`, e);
+        console.debug(
+          `[STREAM-HANDLER] stall watchdog handleError failed:`,
+          retainDiagnosticForLog(e)
+        );
       }
       // Error the pipeline so the downstream reader unblocks. createDisconnect-
       // AwareStream's catch block translates this into buildStreamErrorChunks
@@ -895,13 +906,16 @@ export function pipeWithDisconnect(
       try {
         upstreamTapController?.error(stallError);
       } catch (e) {
-        console.debug(`[STREAM-HANDLER] stall watchdog upstream tap error failed:`, e);
+        console.debug(
+          `[STREAM-HANDLER] stall watchdog upstream tap error failed:`,
+          retainDiagnosticForLog(e)
+        );
       }
       // Abort the underlying fetch so upstream releases the connection.
       try {
         streamController.abort?.();
       } catch (e) {
-        console.debug(`[STREAM-HANDLER] stall watchdog abort failed:`, e);
+        console.debug(`[STREAM-HANDLER] stall watchdog abort failed:`, retainDiagnosticForLog(e));
       }
     }, stallTimeoutMs);
   };

open-sse/utils/streamReadiness.ts

@@ -474,6 +474,8 @@ export async function ensureStreamReadiness(
     provider?: string | null;
     model?: string | null;
     log?: StreamReadinessLogger | null;
+    /** Keep the diagnostic transient while retaining only its existence in logs. */
+    redactUpstreamDiagnosticForLog?: boolean;
   }
 ): Promise<StreamReadinessResult> {
   if (!response.body || options.timeoutMs <= 0) return { ok: true, response };
@@ -568,9 +570,13 @@ export async function ensureStreamReadiness(
         const reason = upstreamDiagnostic
           ? `${classificationReason}: ${upstreamDiagnostic}`
           : classificationReason;
+        const retainedReason =
+          upstreamDiagnostic && options.redactUpstreamDiagnosticForLog
+            ? `${classificationReason}: [upstream diagnostic omitted]`
+            : reason;
         options.log?.warn?.(
           "STREAM",
-          `${reason} (${options.provider || "provider"}/${options.model || "unknown"})`
+          `${retainedReason} (${options.provider || "provider"}/${options.model || "unknown"})`
         );
         return {
           ok: false,

src/lib/guardrails/base.ts

@@ -19,6 +19,8 @@ export interface GuardrailContext {
   sourceFormat?: string | null;
   stream?: boolean;
   targetFormat?: string | null;
+  /** Server-owned bit resolved before any guardrail can replace the raw video carrier. */
+  videoTranscriptSensitive?: boolean;
 }
 
 export interface GuardrailResult<TValue = unknown> {

src/lib/guardrails/registry.ts

@@ -10,6 +10,7 @@ import { VisionBridgeGuardrail } from "./visionBridge";
 import { AudioBridgeGuardrail } from "./audioBridge";
 import { VideoBridgeGuardrail } from "./videoBridge";
 import { CredentialMaskerGuardrail } from "./credentialMasker";
+import { VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER } from "./videoTranscriptLogRedaction";
 
 /**
  * `preCall`/`postCall` may legitimately return nothing — that is the documented
@@ -69,8 +70,29 @@ function coerceDisabledGuardrails(value: unknown) {
     .filter(Boolean);
 }
 
-function getGuardrailLogger(context: GuardrailContext) {
+function getGuardrailLogger(context: GuardrailContext): NonNullable<GuardrailContext["log"]> {
-  return context.log || console;
+  const source = context.log || console;
+  if (context.videoTranscriptSensitive !== true) return source;
+
+  const emit = (
+    level: "debug" | "info" | "warn" | "error",
+    tag: string,
+    _message: string,
+    _meta?: Record<string, unknown>
+  ) => {
+    const target = source[level];
+    if (typeof target !== "function") return;
+    target.call(source, tag, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER, {
+      detail: VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+    });
+  };
+
+  return {
+    debug: (tag, message, meta) => emit("debug", tag, message, meta),
+    info: (tag, message, meta) => emit("info", tag, message, meta),
+    warn: (tag, message, meta) => emit("warn", tag, message, meta),
+    error: (tag, message, meta) => emit("error", tag, message, meta),
+  };
 }
 
 export function resolveDisabledGuardrails({
@@ -135,6 +157,8 @@ export class GuardrailRegistry {
 
   async runPreCallHooks<TPayload = unknown>(payload: TPayload, context: GuardrailContext = {}) {
     const logger = getGuardrailLogger(context);
+    const guardedContext =
+      context.videoTranscriptSensitive === true ? { ...context, log: logger } : context;
     const results: GuardrailExecutionResult[] = [];
     let currentPayload = payload;
 
@@ -151,7 +175,7 @@ export class GuardrailRegistry {
       }
 
       try {
-        const result = asGuardrailResult(await guardrail.preCall(currentPayload, context));
+        const result = asGuardrailResult(await guardrail.preCall(currentPayload, guardedContext));
         const modified = result?.modifiedPayload !== undefined;
         const meta = result?.meta || null;
 
@@ -211,6 +235,8 @@ export class GuardrailRegistry {
 
   async runPostCallHooks<TResponse = unknown>(response: TResponse, context: GuardrailContext = {}) {
     const logger = getGuardrailLogger(context);
+    const guardedContext =
+      context.videoTranscriptSensitive === true ? { ...context, log: logger } : context;
     const results: GuardrailExecutionResult[] = [];
     let currentResponse = response;
 
@@ -227,7 +253,7 @@ export class GuardrailRegistry {
       }
 
       try {
-        const result = asGuardrailResult(await guardrail.postCall(currentResponse, context));
+        const result = asGuardrailResult(await guardrail.postCall(currentResponse, guardedContext));
         const modified = result?.modifiedResponse !== undefined;
         const meta = result?.meta || null;
 

src/lib/guardrails/videoBridge.ts

@@ -18,6 +18,7 @@ import {
   type BridgeCacheStore,
 } from "./modalityBridge/bridgeCache";
 import { recordBridgeUse } from "./modalityBridge/bridgeStats";
+import { fingerprintVideoTranscriptDescription } from "./videoTranscriptLogRedaction";
 import {
   composeVideoFramePrompt,
   describeVideoPart as defaultDescribeVideoPart,
@@ -714,6 +715,16 @@ export class VideoBridgeGuardrail extends BaseGuardrail {
     const videosProcessed = attemptedParts.length - failures;
     const videosReplaced = descriptions.filter((description) => description !== null).length;
     if (videosReplaced === 0) return { block: false };
+    const videoTranscriptDescriptionFingerprints = [
+      ...new Set(
+        descriptions
+          .filter(
+            (description): description is string =>
+              typeof description === "string" && description.includes("transcript[source=")
+          )
+          .map(fingerprintVideoTranscriptDescription)
+      ),
+    ].sort();
 
     return {
       block: false,
@@ -731,6 +742,9 @@ export class VideoBridgeGuardrail extends BaseGuardrail {
         focusWindowsApplied,
         focusHintsApplied,
         transcriptCuesApplied,
+        ...(videoTranscriptDescriptionFingerprints.length > 0
+          ? { videoTranscriptDescriptionFingerprints }
+          : {}),
         contactSheetsUsed,
         audioFusionRuns,
         audioFusionPartials,

src/lib/guardrails/videoTranscriptLogRedaction.ts

@@ -0,0 +1,512 @@
+import { createHash } from "node:crypto";
+
+export const VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER = "[omitted: video transcript]";
+export const VIDEO_TRANSCRIPT_REDACTION_PIPELINE_KEY = "_omnirouteVideoTranscriptRedacted" as const;
+
+const VIDEO_TRANSCRIPT_PAYLOAD_KEYS = new Set(["audioTranscript", "transcript"]);
+const VIDEO_TRANSCRIPT_CARRIER_KEYS = new Set(["input_video", "source", "url", "video_url"]);
+const VIDEO_BRIDGE_DESCRIPTION_PREFIX = "[Video description:";
+// Security bound for the retained copy, deliberately above the default downstream log depth (20).
+// Crossing it is unknown-as-sensitive: callers must discard the partial copy instead of leaking it.
+const MAX_VIDEO_TRANSCRIPT_LOG_SECURITY_DEPTH = 32;
+// Aggregate bound across the whole retained graph, not a per-object width limit. The partial copy
+// is discarded on overflow so attacker-controlled breadth cannot produce a large unsafe artifact.
+const MAX_VIDEO_TRANSCRIPT_LOG_SECURITY_ENTRIES = 10_000;
+const VIDEO_TRANSCRIPT_CUE_PREFIX = "transcript[source=";
+const VIDEO_TRANSCRIPT_CUE_SOURCES = ["audio-bridge", "client", "embedded"] as const;
+const MAX_VIDEO_TRANSCRIPT_CUE_METADATA_CODE_UNITS = 1024;
+const VIDEO_TRANSCRIPT_DESCRIPTION_FINGERPRINT_RE = /^sha256:([a-f0-9]{64}):(\d{1,8})$/;
+const MAX_TRUSTED_DESCRIPTION_CODE_UNITS = 16 * 1024 * 1024;
+const MAX_TRUSTED_DESCRIPTION_IDENTITIES = 64;
+const MAX_TRUSTED_DESCRIPTION_HASH_CODE_UNITS = 64 * 1024 * 1024;
+const MAX_TRUSTED_DESCRIPTION_PREFIX_OCCURRENCES = 128;
+const MAX_TRUSTED_DESCRIPTION_CANDIDATE_HASHES = 512;
+
+type JsonRecord = Record<string, unknown>;
+
+export interface VideoTranscriptLogContext {
+  /** SHA-256 identities emitted only for descriptions produced by the Video Bridge guardrail. */
+  trustedDescriptionFingerprints?: readonly string[];
+}
+
+export interface VideoTranscriptLogSensitivityInput {
+  /** Original client body, before any guardrail is allowed to replace or remove media parts. */
+  rawRequestBody?: unknown;
+  /** Current request body after guardrail processing. */
+  processedBody?: unknown;
+  /** Validated identities emitted by a successful Video Bridge rewrite. */
+  trustedDescriptionFingerprints?: readonly string[];
+}
+
+export function fingerprintVideoTranscriptDescription(value: string): string {
+  return `sha256:${createHash("sha256").update(value).digest("hex")}:${value.length}`;
+}
+
+/** Keep raw provider text transient while replacing every retained/logged copy. */
+export function redactVideoTranscriptSensitiveText(value: string, sensitive: boolean): string {
+  return sensitive ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER : value;
+}
+
+type WalkFrame = {
+  directCarrier: boolean;
+  depth: number;
+  value: object;
+};
+
+type CloneFrame = WalkFrame & {
+  target: JsonRecord | unknown[];
+};
+
+type CloneMemo = {
+  carrier?: JsonRecord | unknown[];
+  ordinary?: JsonRecord | unknown[];
+};
+
+type TrustedDescriptionIdentity = {
+  fingerprint: string;
+  length: number;
+};
+
+type TrustedDescriptionRange = {
+  end: number;
+  start: number;
+};
+
+type VideoTranscriptCuePrefix = { kind: "valid"; textStart: number } | { kind: "malformed" };
+
+function parseTrustedDescriptionIdentity(value: unknown): TrustedDescriptionIdentity | null {
+  if (typeof value !== "string") return null;
+  const match = VIDEO_TRANSCRIPT_DESCRIPTION_FINGERPRINT_RE.exec(value);
+  if (!match) return null;
+  const length = Number(match[2]);
+  if (!Number.isSafeInteger(length) || length <= 0 || length > MAX_TRUSTED_DESCRIPTION_CODE_UNITS) {
+    return null;
+  }
+  return { fingerprint: value, length };
+}
+
+function isJsonRecord(value: unknown): value is JsonRecord {
+  return Boolean(value && typeof value === "object" && !Array.isArray(value));
+}
+
+function isWhitespaceCodeUnit(value: string, index: number): boolean {
+  const codeUnit = value[index];
+  return codeUnit !== undefined && codeUnit.trim().length === 0;
+}
+
+function findNextVideoTranscriptCuePrefix(
+  value: string,
+  fromIndex: number,
+  end: number
+): VideoTranscriptCuePrefix | null {
+  const start = value.indexOf(VIDEO_TRANSCRIPT_CUE_PREFIX, fromIndex);
+  if (start < 0 || start >= end) return null;
+
+  const sourceStart = start + VIDEO_TRANSCRIPT_CUE_PREFIX.length;
+  const source = VIDEO_TRANSCRIPT_CUE_SOURCES.find((candidate) =>
+    value.startsWith(candidate, sourceStart)
+  );
+  if (!source) return { kind: "malformed" };
+
+  const metadataStart = sourceStart + source.length;
+  const metadataLimit = Math.min(
+    end,
+    metadataStart + MAX_VIDEO_TRANSCRIPT_CUE_METADATA_CODE_UNITS + 1
+  );
+  let metadataEnd = -1;
+  for (let cursor = metadataStart; cursor < metadataLimit; cursor += 1) {
+    if (value[cursor] === "\r" || value[cursor] === "\n") break;
+    if (value[cursor] === "]") {
+      metadataEnd = cursor;
+      break;
+    }
+  }
+  if (metadataEnd < metadataStart) return { kind: "malformed" };
+
+  let cursor = metadataEnd + 1;
+  const whitespaceStart = cursor;
+  while (cursor < end && isWhitespaceCodeUnit(value, cursor)) cursor += 1;
+  if (
+    cursor === whitespaceStart ||
+    cursor + "text=".length > end ||
+    !value.startsWith("text=", cursor)
+  ) {
+    return { kind: "malformed" };
+  }
+  return { kind: "valid", textStart: cursor + "text=".length };
+}
+
+function findQuotedStringEnd(value: string, quoteStart: number, end: number): number | null {
+  if (value[quoteStart] !== '"') return null;
+  let cursor = quoteStart + 1;
+  while (cursor < end) {
+    if (value[cursor] === '"') return cursor + 1;
+    if (value[cursor] === "\\") {
+      cursor += 2;
+    } else {
+      cursor += 1;
+    }
+  }
+  return null;
+}
+
+function redactSerializedVideoTranscriptCues(value: string): string {
+  const pieces: string[] = [];
+  let cursor = 0;
+  let searchFrom = 0;
+  while (searchFrom < value.length) {
+    const cue = findNextVideoTranscriptCuePrefix(value, searchFrom, value.length);
+    if (!cue) break;
+    if (cue.kind === "malformed") return VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+    const quotedEnd = findQuotedStringEnd(value, cue.textStart, value.length);
+    if (quotedEnd === null) return VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+    pieces.push(
+      value.slice(cursor, cue.textStart),
+      JSON.stringify(VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER)
+    );
+    cursor = quotedEnd;
+    searchFrom = quotedEnd;
+  }
+  // This parser is called only for a description whose exact fingerprint was emitted by the
+  // Video Bridge after applying at least one transcript cue. No parseable cue therefore means
+  // malformed trusted data: omit the whole retained range instead of returning it verbatim.
+  if (pieces.length === 0) return VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+  pieces.push(value.slice(cursor));
+  return pieces.join("");
+}
+
+/** Accept only identities emitted by a successful server-side Video Bridge rewrite. */
+export function extractVideoTranscriptDescriptionFingerprints(results: unknown): string[] {
+  if (!Array.isArray(results)) return [];
+  const fingerprints = new Set<string>();
+  for (const result of results) {
+    if (!isJsonRecord(result) || result.guardrail !== "video-bridge" || result.modified !== true) {
+      continue;
+    }
+    const meta = isJsonRecord(result.meta) ? result.meta : null;
+    if (
+      !meta ||
+      typeof meta.transcriptCuesApplied !== "number" ||
+      meta.transcriptCuesApplied <= 0 ||
+      !Array.isArray(meta.videoTranscriptDescriptionFingerprints)
+    ) {
+      continue;
+    }
+    for (const fingerprint of meta.videoTranscriptDescriptionFingerprints) {
+      const identity = parseTrustedDescriptionIdentity(fingerprint);
+      if (identity) fingerprints.add(identity.fingerprint);
+    }
+  }
+  return [...fingerprints].sort();
+}
+
+function isRecognizedVideoPart(record: JsonRecord): boolean {
+  const type = typeof record.type === "string" ? record.type : undefined;
+  // Explicit modality tags remain security carriers even when the media URL/data is malformed.
+  // Validation can reject them later, but malformed input must never bypass retained-log policy.
+  if (
+    type === "input_video" ||
+    type === "video" ||
+    type === "video_source" ||
+    type === "video_url"
+  ) {
+    return true;
+  }
+  if (Object.hasOwn(record, "input_video") || Object.hasOwn(record, "video_url")) return true;
+
+  const source = isJsonRecord(record.source) ? record.source : undefined;
+  if (!source) return false;
+  const videoMediaType =
+    typeof source.media_type === "string" && source.media_type.toLowerCase().startsWith("video/");
+  return videoMediaType;
+}
+
+function findTrustedDescriptionRanges(
+  value: string,
+  context: VideoTranscriptLogContext
+): TrustedDescriptionRange[] | null {
+  if (!value.includes(VIDEO_BRIDGE_DESCRIPTION_PREFIX)) return [];
+  const identities = context.trustedDescriptionFingerprints
+    ?.map(parseTrustedDescriptionIdentity)
+    .filter((identity): identity is TrustedDescriptionIdentity => identity !== null);
+  if (!identities || identities.length === 0) return [];
+  // A successful request can only produce a bounded number of video descriptions.
+  // If internal metadata violates that contract, omit the whole retained string
+  // rather than risk leaking a cue or performing attacker-amplified hashing.
+  if (identities.length > MAX_TRUSTED_DESCRIPTION_IDENTITIES) return null;
+
+  const ranges: TrustedDescriptionRange[] = [];
+  let candidateHashes = 0;
+  let hashBudget = 0;
+  let prefixOccurrences = 0;
+  let start = value.indexOf(VIDEO_BRIDGE_DESCRIPTION_PREFIX);
+  while (start >= 0) {
+    prefixOccurrences += 1;
+    // Bound synchronous work independently of byte length: many tiny forged
+    // prefixes would otherwise amplify hashing and block the event loop. A
+    // retained copy fails closed while the live request remains untouched.
+    if (prefixOccurrences > MAX_TRUSTED_DESCRIPTION_PREFIX_OCCURRENCES) return null;
+    for (const identity of identities) {
+      const end = start + identity.length;
+      if (end > value.length) continue;
+      candidateHashes += 1;
+      if (candidateHashes > MAX_TRUSTED_DESCRIPTION_CANDIDATE_HASHES) return null;
+      hashBudget += identity.length;
+      if (hashBudget > MAX_TRUSTED_DESCRIPTION_HASH_CODE_UNITS) return null;
+      const candidate = value.slice(start, end);
+      if (fingerprintVideoTranscriptDescription(candidate) === identity.fingerprint) {
+        ranges.push({ end, start });
+      }
+    }
+    start = value.indexOf(VIDEO_BRIDGE_DESCRIPTION_PREFIX, start + 1);
+  }
+
+  ranges.sort((left, right) => left.start - right.start || left.end - right.end);
+  const merged: TrustedDescriptionRange[] = [];
+  for (const range of ranges) {
+    const previous = merged.at(-1);
+    if (previous && range.start < previous.end) {
+      previous.end = Math.max(previous.end, range.end);
+    } else {
+      merged.push({ ...range });
+    }
+  }
+  return merged;
+}
+
+export function omitVideoTranscriptFromLogString(
+  value: string,
+  context: VideoTranscriptLogContext = {}
+): string {
+  const ranges = findTrustedDescriptionRanges(value, context);
+  if (ranges === null) return VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+  if (ranges.length === 0) return value;
+  let result = "";
+  let cursor = 0;
+  for (const range of ranges) {
+    result += value.slice(cursor, range.start);
+    result += redactSerializedVideoTranscriptCues(value.slice(range.start, range.end));
+    cursor = range.end;
+  }
+  return result + value.slice(cursor);
+}
+
+/**
+ * Remove complete trusted Video Bridge descriptions before Memory extraction.
+ * Logs may retain the non-transcript portion of a description, but durable
+ * user facts must come only from caller text, never from media-derived prose.
+ * Exact server-issued fingerprints let adjacent caller text survive even when
+ * both appear in the same string. Unknown/over-budget input fails closed.
+ */
+export function omitVideoTranscriptDerivedTextForMemory(
+  value: string,
+  context: VideoTranscriptLogContext = {}
+): string {
+  try {
+    const ranges = findTrustedDescriptionRanges(value, context);
+    if (ranges === null) return "";
+    if (ranges.length === 0) return value;
+
+    let result = "";
+    let cursor = 0;
+    for (const range of ranges) {
+      result += value.slice(cursor, range.start);
+      cursor = range.end;
+    }
+    return result + value.slice(cursor);
+  } catch {
+    return "";
+  }
+}
+
+function fieldIsTranscript(key: string, carrier: boolean): boolean {
+  return carrier && VIDEO_TRANSCRIPT_PAYLOAD_KEYS.has(key);
+}
+
+function childIsDirectCarrier(key: string, carrierParent: boolean, value: unknown): boolean {
+  return carrierParent && VIDEO_TRANSCRIPT_CARRIER_KEYS.has(key) && isJsonRecord(value);
+}
+
+function* enumerableEntries(value: object): Generator<readonly [string, unknown]> {
+  if (Array.isArray(value)) {
+    for (let index = 0; index < value.length; index += 1) {
+      yield [String(index), value[index]] as const;
+    }
+    return;
+  }
+
+  const record = value as JsonRecord;
+  for (const key in record) {
+    if (Object.hasOwn(record, key)) yield [key, record[key]] as const;
+  }
+}
+
+function walkContains(root: unknown, context: VideoTranscriptLogContext): boolean {
+  const stack: Array<WalkFrame | { directCarrier: boolean; depth: number; value: string }> = [];
+  if (typeof root === "function") return true;
+  if (typeof root === "string") stack.push({ directCarrier: false, depth: 0, value: root });
+  else if (root && typeof root === "object" && !ArrayBuffer.isView(root)) {
+    stack.push({ directCarrier: false, depth: 0, value: root });
+  }
+  const seen = new WeakMap<object, number>();
+  let scannedEntries = 0;
+
+  while (stack.length > 0) {
+    const frame = stack.pop()!;
+    if (frame.depth > MAX_VIDEO_TRANSCRIPT_LOG_SECURITY_DEPTH) return true;
+    if (typeof frame.value === "string") {
+      const stringValue = frame.value;
+      const ranges = findTrustedDescriptionRanges(stringValue, context);
+      if (ranges === null) return true;
+      if (
+        ranges.some(
+          (range) => findNextVideoTranscriptCuePrefix(stringValue, range.start, range.end) !== null
+        )
+      ) {
+        return true;
+      }
+      continue;
+    }
+
+    const record = Array.isArray(frame.value) ? null : (frame.value as JsonRecord);
+    const recognized = record ? isRecognizedVideoPart(record) : false;
+    const carrier = frame.directCarrier || recognized;
+    const seenBit = carrier ? 2 : 1;
+    const previousBits = seen.get(frame.value) ?? 0;
+    if ((previousBits & seenBit) !== 0) continue;
+    seen.set(frame.value, previousBits | seenBit);
+
+    for (const [key, value] of enumerableEntries(frame.value)) {
+      scannedEntries += 1;
+      if (scannedEntries > MAX_VIDEO_TRANSCRIPT_LOG_SECURITY_ENTRIES) return true;
+      if (fieldIsTranscript(key, carrier)) {
+        return true;
+      }
+      // Functions are not valid JSON request values. In particular, an enumerable toJSON
+      // function could replace the retained clone with attacker-selected transcript content.
+      if (typeof value === "function") return true;
+      if (typeof value === "string") {
+        stack.push({
+          directCarrier: fieldIsTranscript(key, carrier),
+          depth: frame.depth + 1,
+          value,
+        });
+      } else if (value && typeof value === "object" && !ArrayBuffer.isView(value)) {
+        stack.push({
+          directCarrier: childIsDirectCarrier(key, carrier, value),
+          depth: frame.depth + 1,
+          value,
+        });
+      }
+    }
+  }
+  return false;
+}
+
+/** Detect transcript data only in recognized video carriers or a trusted bridge description. */
+export function containsVideoTranscriptForLog(
+  value: unknown,
+  context: VideoTranscriptLogContext = {}
+): boolean {
+  try {
+    return walkContains(value, context);
+  } catch {
+    return true;
+  }
+}
+
+/** Resolve sensitivity once from both sides of the guardrail boundary. */
+export function resolveVideoTranscriptLogSensitivity({
+  rawRequestBody,
+  processedBody,
+  trustedDescriptionFingerprints = [],
+}: VideoTranscriptLogSensitivityInput): boolean {
+  const hasTrustedDescription = trustedDescriptionFingerprints.some(
+    (fingerprint) => parseTrustedDescriptionIdentity(fingerprint) !== null
+  );
+  return (
+    hasTrustedDescription ||
+    containsVideoTranscriptForLog(rawRequestBody) ||
+    containsVideoTranscriptForLog(processedBody)
+  );
+}
+
+function prepareClone(
+  value: unknown,
+  depth: number,
+  directCarrier: boolean,
+  context: VideoTranscriptLogContext,
+  memo: WeakMap<object, CloneMemo>
+): { frame?: CloneFrame; limitExceeded?: boolean; value: unknown } {
+  if (depth > MAX_VIDEO_TRANSCRIPT_LOG_SECURITY_DEPTH) {
+    return { limitExceeded: true, value: VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER };
+  }
+  if (typeof value === "string") {
+    return { value: omitVideoTranscriptFromLogString(value, context) };
+  }
+  if (typeof value === "function") return { value: "[Function]" };
+  if (!value || typeof value !== "object") return { value };
+  if (ArrayBuffer.isView(value)) return { value: `[binary ${value.byteLength} bytes]` };
+
+  const record = Array.isArray(value) ? null : (value as JsonRecord);
+  const carrier = directCarrier || Boolean(record && isRecognizedVideoPart(record));
+  const memoKey = carrier ? "carrier" : "ordinary";
+  const previous = memo.get(value)?.[memoKey];
+  if (previous) return { value: "[Circular]" };
+
+  const target: JsonRecord | unknown[] = Array.isArray(value) ? [] : {};
+  const entry = memo.get(value) ?? {};
+  entry[memoKey] = target;
+  memo.set(value, entry);
+  return { frame: { directCarrier: carrier, depth, target, value }, value: target };
+}
+
+function omitVideoTranscriptForLogUnsafe(
+  payload: unknown,
+  context: VideoTranscriptLogContext
+): unknown {
+  const memo = new WeakMap<object, CloneMemo>();
+  const prepared = prepareClone(payload, 0, false, context, memo);
+  if (prepared.limitExceeded) return VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+  if (!prepared.frame) return prepared.value;
+  const stack = [prepared.frame];
+  let clonedEntries = 0;
+
+  while (stack.length > 0) {
+    const frame = stack.pop()!;
+    for (const [key, value] of enumerableEntries(frame.value)) {
+      clonedEntries += 1;
+      if (clonedEntries > MAX_VIDEO_TRANSCRIPT_LOG_SECURITY_ENTRIES) {
+        return VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+      }
+      const targetKey: string | number = Array.isArray(frame.target) ? Number(key) : key;
+      if (fieldIsTranscript(key, frame.directCarrier)) {
+        (frame.target as JsonRecord)[targetKey] = VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+        continue;
+      }
+      const child = prepareClone(
+        value,
+        frame.depth + 1,
+        childIsDirectCarrier(key, frame.directCarrier, value),
+        context,
+        memo
+      );
+      if (child.limitExceeded) return VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+      (frame.target as JsonRecord)[targetKey] = child.value;
+      if (child.frame) stack.push(child.frame);
+    }
+  }
+  return prepared.value;
+}
+
+/** Remove Video Bridge source fields and trusted generated segments from a bounded log copy. */
+export function omitVideoTranscriptForLog(
+  payload: unknown,
+  context: VideoTranscriptLogContext = {}
+): unknown {
+  try {
+    return omitVideoTranscriptForLogUnsafe(payload, context);
+  } catch {
+    return VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER;
+  }
+}

src/lib/logPayloads.ts

@@ -1,4 +1,10 @@
 import { sanitizePII } from "./piiSanitizer";
+import { omitVideoTranscriptForLog } from "./guardrails/videoTranscriptLogRedaction";
+
+export {
+  omitVideoTranscriptForLog,
+  VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+} from "./guardrails/videoTranscriptLogRedaction";
 
 const SENSITIVE_KEYS = new Set([
   "api_key",
@@ -162,7 +168,8 @@ export function sanitizePayloadPII(payload: unknown): unknown {
 export function protectPayloadForLog(payload: unknown): unknown {
   if (payload === null || payload === undefined) return null;
   const normalized = normalizePayloadForLog(payload);
-  const reasoningOmitted = omitEncryptedReasoningForLog(normalized);
+  const transcriptOmitted = omitVideoTranscriptForLog(normalized);
+  const reasoningOmitted = omitEncryptedReasoningForLog(transcriptOmitted);
   const piiSanitized = sanitizePayloadPII(reasoningOmitted);
   return redactPayload(piiSanitized);
 }

src/lib/plugins/hooks.ts

@@ -8,9 +8,25 @@
  */
 
 import { logger } from "../../../open-sse/utils/logger.ts";
+import { redactVideoTranscriptSensitiveText } from "../guardrails/videoTranscriptLogRedaction.ts";
 
 const log = logger("PLUGIN_HOOKS");
 
+function isVideoTranscriptSensitiveContext(value: unknown): boolean {
+  return Boolean(
+    value &&
+    typeof value === "object" &&
+    (value as Record<string, unknown>).videoTranscriptSensitive === true
+  );
+}
+
+function retainedHookError(error: unknown, sensitive: boolean): string {
+  return redactVideoTranscriptSensitiveText(
+    error instanceof Error ? error.message : String(error),
+    sensitive
+  );
+}
+
 // ── Types ──
 
 export type BlockingHookResult = {
@@ -143,6 +159,7 @@ export function unregisterHook(event: string, pluginName: string): void {
 export async function emitHook(event: string, payload: unknown): Promise<void> {
   const list = hooks.get(event);
   if (!list || list.length === 0) return;
+  const videoTranscriptSensitive = isVideoTranscriptSensitiveContext(payload);
 
   for (const reg of list) {
     if (isRateLimited(reg.pluginName)) {
@@ -152,11 +169,10 @@ export async function emitHook(event: string, payload: unknown): Promise<void> {
     try {
       await reg.handler(payload);
     } catch (err: unknown) {
-      const message = err instanceof Error ? err.message : String(err);
       log.error("hook.handler_error", {
         event,
         pluginName: reg.pluginName,
-        error: message,
+        error: retainedHookError(err, videoTranscriptSensitive),
       });
     }
   }
@@ -178,6 +194,7 @@ export async function emitHookBlocking(
 }> {
   const list = hooks.get(event) || [];
   const ctx = (payload || {}) as Record<string, unknown>;
+  const videoTranscriptSensitive = isVideoTranscriptSensitiveContext(ctx);
   let mergedBody: unknown = ctx.body;
   let mergedMetadata: Record<string, unknown> = (ctx.metadata as Record<string, unknown>) || {};
 
@@ -209,11 +226,10 @@ export async function emitHookBlocking(
         }
       }
     } catch (err: unknown) {
-      const message = err instanceof Error ? err.message : String(err);
       log.error("hook.blocking_handler_error", {
         event,
         pluginName: reg.pluginName,
-        error: message,
+        error: retainedHookError(err, videoTranscriptSensitive),
       });
     }
   }
@@ -234,6 +250,8 @@ export interface PluginContext {
    *  client (trace ids, correlation ids, session markers). */
   headers?: Record<string, string | string[] | undefined>;
   metadata: Record<string, unknown>;
+  /** Server-owned retention bit; plugins still receive the live body as trusted processors. */
+  videoTranscriptSensitive?: boolean;
 }
 
 export interface PluginResult {
@@ -333,6 +351,7 @@ export async function runOnRequest(ctx: PluginContext): Promise<PluginResult> {
 export async function runOnResponse(ctx: PluginContext, response: unknown): Promise<unknown> {
   let currentResponse = response;
   const list = hooks.get("onResponse") || [];
+  const videoTranscriptSensitive = ctx.videoTranscriptSensitive === true;
   for (const reg of list) {
     try {
       const result = await reg.handler({ ...ctx, response: currentResponse });
@@ -345,8 +364,10 @@ export async function runOnResponse(ctx: PluginContext, response: unknown): Prom
         currentResponse = (result as { response: unknown }).response;
       }
     } catch (err: unknown) {
-      const message = err instanceof Error ? err.message : String(err);
+      log.error("hook.response_handler_error", {
-      log.error("hook.response_handler_error", { pluginName: reg.pluginName, error: message });
+        pluginName: reg.pluginName,
+        error: retainedHookError(err, videoTranscriptSensitive),
+      });
     }
   }
   return currentResponse;

src/lib/usage/pendingRequestScope.ts

@@ -11,16 +11,32 @@ export type PendingRequestScope = {
   model: string;
   provider: string;
   connectionId: string | null;
+  videoTranscriptSensitive?: boolean;
+  videoTranscriptDescriptionFingerprints?: readonly string[];
 };
 
 export function updatePendingScope(scope: PendingRequestScope, metadata: PendingRequestMetadata) {
-  if (!updatePendingRequestById(scope.id || null, metadata)) {
+  const protectedMetadata = {
-    updatePendingRequest(scope.model, scope.provider, scope.connectionId, metadata);
+    ...metadata,
+    ...(scope.videoTranscriptSensitive ? { videoTranscriptSensitive: true } : {}),
+    ...(scope.videoTranscriptDescriptionFingerprints?.length
+      ? { videoTranscriptDescriptionFingerprints: scope.videoTranscriptDescriptionFingerprints }
+      : {}),
+  };
+  if (!updatePendingRequestById(scope.id || null, protectedMetadata)) {
+    updatePendingRequest(scope.model, scope.provider, scope.connectionId, protectedMetadata);
   }
 }
 
 export function finalizePendingScope(scope: PendingRequestScope, metadata: PendingRequestMetadata) {
-  if (!finalizePendingRequestById(scope.id, metadata)) {
+  const protectedMetadata = {
-    finalizePendingRequest(scope.model, scope.provider, scope.connectionId, metadata);
+    ...metadata,
+    ...(scope.videoTranscriptSensitive ? { videoTranscriptSensitive: true } : {}),
+    ...(scope.videoTranscriptDescriptionFingerprints?.length
+      ? { videoTranscriptDescriptionFingerprints: scope.videoTranscriptDescriptionFingerprints }
+      : {}),
+  };
+  if (!finalizePendingRequestById(scope.id, protectedMetadata)) {
+    finalizePendingRequest(scope.model, scope.provider, scope.connectionId, protectedMetadata);
   }
 }

src/lib/usage/usageHistory.ts

@@ -8,6 +8,11 @@
  */
 
 import { getDbInstance } from "../db/core";
+import {
+  omitVideoTranscriptForLog,
+  type VideoTranscriptLogContext,
+  VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+} from "../guardrails/videoTranscriptLogRedaction";
 import { protectPayloadForLog } from "../logPayloads";
 import {
   resolveOrphanedUsageAccountIdentity,
@@ -55,6 +60,10 @@ export type PendingRequestMetadata = {
   stageUpdatedAt?: number | null;
   correlationId?: string | null;
   sessionTag?: string | null;
+  /** Trusted request state; consumed during normalization and never retained. */
+  videoTranscriptSensitive?: boolean;
+  /** Exact SHA-256 identities of generated Video descriptions; never retained. */
+  videoTranscriptDescriptionFingerprints?: readonly string[];
 };
 export type PendingRequestDetail = {
   id: string;
@@ -88,6 +97,14 @@ function normalizePendingMetadata(metadata?: PendingRequestMetadata): PendingReq
   if (!metadata) return {};
 
   const normalized: PendingRequestMetadata = {};
+  const transcriptSensitive = metadata.videoTranscriptSensitive === true;
+  const descriptionContext: VideoTranscriptLogContext = {
+    trustedDescriptionFingerprints: metadata.videoTranscriptDescriptionFingerprints ?? [],
+  };
+  const protectRequest = (value: unknown, logContext: VideoTranscriptLogContext = {}): unknown =>
+    truncatePendingPreview(protectPayloadForLog(omitVideoTranscriptForLog(value, logContext)));
+  const protectResponse = (value: unknown): unknown =>
+    transcriptSensitive ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER : protectRequest(value);
 
   if (metadata.clientEndpoint !== undefined) {
     normalized.clientEndpoint = toStringOrNull(metadata.clientEndpoint) || null;
@@ -106,29 +123,25 @@ function normalizePendingMetadata(metadata?: PendingRequestMetadata): PendingReq
         : null;
   }
   if (metadata.clientRequest !== undefined) {
-    normalized.clientRequest = truncatePendingPreview(protectPayloadForLog(metadata.clientRequest));
+    normalized.clientRequest = protectRequest(metadata.clientRequest, {});
   }
   if (metadata.providerRequest !== undefined) {
-    normalized.providerRequest = truncatePendingPreview(
+    normalized.providerRequest = protectRequest(metadata.providerRequest, descriptionContext);
-      protectPayloadForLog(metadata.providerRequest)
-    );
   }
   if (metadata.providerResponse !== undefined) {
-    normalized.providerResponse = truncatePendingPreview(
+    normalized.providerResponse = protectResponse(metadata.providerResponse);
-      protectPayloadForLog(metadata.providerResponse)
-    );
   }
   if (metadata.clientResponse !== undefined) {
-    normalized.clientResponse = truncatePendingPreview(
+    normalized.clientResponse = protectResponse(metadata.clientResponse);
-      protectPayloadForLog(metadata.clientResponse)
-    );
   }
   if (metadata.status !== undefined) {
     const status = Number(metadata.status);
     normalized.status = Number.isFinite(status) ? status : null;
   }
   if (metadata.error !== undefined) {
-    normalized.error = toStringOrNull(metadata.error) || null;
+    normalized.error = transcriptSensitive
+      ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER
+      : toStringOrNull(metadata.error) || null;
   }
   if (metadata.errorCode !== undefined) {
     normalized.errorCode = toStringOrNull(metadata.errorCode) || null;

src/sse/handlers/chat.ts

@@ -6,6 +6,11 @@ export { buildClientRawRequest, resolveDispatchClientRawRequest };
 import { normalizeReasoningRequest } from "@/shared/reasoning/effortStandardization";
 import { isDetailedLoggingEnabled } from "@/lib/db/detailedLogs";
 import { resolvePreviousResponseState } from "@/lib/db/responsesContinuationStore";
+import {
+  extractVideoTranscriptDescriptionFingerprints,
+  redactVideoTranscriptSensitiveText,
+  resolveVideoTranscriptLogSensitivity,
+} from "@/lib/guardrails/videoTranscriptLogRedaction";
 import { normalizeResponsesPreviousResponseIdMode } from "@omniroute/open-sse/utils/responsesStatePolicy.ts";
 import { FORMATS } from "@omniroute/open-sse/translator/formats.ts";
 import { resolveRoutingModel, RoutingModelOps } from "./resolveRoutingModel";
@@ -102,6 +107,7 @@ import { resolveConversationId } from "@omniroute/open-sse/services/conversation
 import {
   isAntigravityMissingProjectError,
   isProviderBreakerFailureStatus,
+  resolveDailyQuotaModelViews,
   resolveStreamReadinessClassificationError,
   shouldTripProviderBreakerForResult,
 } from "./chatPredicates";
@@ -690,6 +696,10 @@ async function handleChatImplementation(
   clientRawRequest = chatAdmission.resolveClientRawAfterAdmission(clientRawRequest, () =>
     deferredClientRawBody.withClientBody((clientBody) => buildClientRawRequest(request, clientBody))
   );
+  let videoTranscriptSensitive = resolveVideoTranscriptLogSensitivity({
+    rawRequestBody: clientRawRequest?.body,
+    processedBody: body,
+  });
 
   // Guardrail pre-call pipeline — prompt injection, PII masking, and future custom rules.
   telemetry.startPhase("validate");
@@ -707,11 +717,15 @@ async function handleChatImplementation(
     model: modelStr,
     signal: request.signal,
     stream: body?.stream === true,
+    videoTranscriptSensitive,
   });
   if (preCallGuardrails.blocked) {
     log.warn("GUARDRAIL", "Request blocked during pre-call guardrails", {
       guardrail: preCallGuardrails.guardrail,
-      message: preCallGuardrails.message,
+      message: redactVideoTranscriptSensitiveText(
+        preCallGuardrails.message || "Request rejected by guardrail",
+        videoTranscriptSensitive
+      ),
     });
     return errorResponse(
       HTTP_STATUS.BAD_REQUEST,
@@ -722,6 +736,14 @@ async function handleChatImplementation(
   const modelBeforeGuardrails =
     typeof body?.model === "string" && body.model.length > 0 ? body.model : modelStr;
   body = preCallGuardrails.payload;
+  const videoTranscriptDescriptionFingerprints = extractVideoTranscriptDescriptionFingerprints(
+    preCallGuardrails.results
+  );
+  videoTranscriptSensitive = resolveVideoTranscriptLogSensitivity({
+    rawRequestBody: clientRawRequest?.body,
+    processedBody: body,
+    trustedDescriptionFingerprints: videoTranscriptDescriptionFingerprints,
+  });
   ({ body, modelStr } = await RoutingModelOps.reconcileGuardrailReroute({
     body,
     modelBeforeGuardrails,
@@ -757,7 +779,10 @@ async function handleChatImplementation(
     // already treat null/undefined as "untracked" (see withConversationId).
     log.warn("CHAT", "resolveConversationId failed, continuing without conversation tracking", {
       correlationId: reqId,
-      error: error instanceof Error ? error.message : String(error),
+      error: redactVideoTranscriptSensitiveText(
+        error instanceof Error ? error.message : String(error),
+        videoTranscriptSensitive
+      ),
     });
   }
 
@@ -1041,6 +1066,8 @@ async function handleChatImplementation(
       sourceFormat,
       endpointPath: new URL(request.url).pathname,
       requestHeaders: request.headers,
+      videoTranscriptSensitive,
+      videoTranscriptDescriptionFingerprints,
       clientManagedResponsesContext:
         sourceFormat === "openai-responses" &&
         new URL(request.url).pathname.split("/").includes("responses") &&
@@ -1094,6 +1121,8 @@ async function handleChatImplementation(
             reasoningIntent,
             reasoningRequestTags: requestRoutingTags.tags,
             managedLease,
+            videoTranscriptSensitive,
+            videoTranscriptDescriptionFingerprints,
             // #7360 follow-up: without this, a target dispatch abandoned by
             // targetTimeoutRunner.ts's per-target timeout (comboTargetTimeoutMs)
             // never learns it was abandoned — it only watches the ORIGINAL
@@ -1164,6 +1193,8 @@ async function handleChatImplementation(
             forceLiveComboTest: isComboLiveTest,
             conversationId,
             managedLease,
+            videoTranscriptSensitive,
+            videoTranscriptDescriptionFingerprints,
           },
           combo.strategy,
           true
@@ -1181,7 +1212,13 @@ async function handleChatImplementation(
           `Global fallback ${fallbackModel} also failed (${fallbackResponse.status})`
         );
       } catch (err: any) {
-        log.warn("GLOBAL_FALLBACK", `Global fallback error: ${err?.message || "unknown"}`);
+        log.warn(
+          "GLOBAL_FALLBACK",
+          `Global fallback error: ${redactVideoTranscriptSensitiveText(
+            err?.message || "unknown",
+            videoTranscriptSensitive
+          )}`
+        );
       }
     }
     // ─────────────────────────────────────────────────────────────────────────
@@ -1209,6 +1246,7 @@ async function handleChatImplementation(
           sessionTag: conversationId,
           startTime: telemetry?.startTime,
           requestBody: clientRawRequest?.body ?? null,
+          videoTranscriptSensitive,
         });
       } catch {}
     }
@@ -1257,6 +1295,8 @@ async function handleChatImplementation(
       reasoningIntent,
       reasoningRequestTags: requestRoutingTags.tags,
       managedLease,
+      videoTranscriptSensitive,
+      videoTranscriptDescriptionFingerprints,
     },
     null,
     false
@@ -1305,6 +1345,10 @@ async function handleSingleModelChat(
     reasoningRequestTags?: string[];
     reasoningTransportFallback?: "skip" | "drop";
     managedLease?: ManagedLeaseDispatchContext | null;
+    /** Sensitive carrier/result state derived structurally, never from caller prose. */
+    videoTranscriptSensitive?: boolean;
+    /** Exact SHA-256 identities of transcript descriptions generated by the guardrail. */
+    videoTranscriptDescriptionFingerprints?: readonly string[];
     /**
      * Per-target abort signal from combo.ts's targetTimeoutRunner
      * (comboTargetTimeoutMs) — see the #7360 follow-up comment at the
@@ -1316,6 +1360,9 @@ async function handleSingleModelChat(
   comboStrategy: string | null = null,
   isCombo: boolean = false
 ) {
+  const videoTranscriptDescriptionFingerprints =
+    runtimeOptions.videoTranscriptDescriptionFingerprints ?? [];
+
   // 1. Resolve model → provider/model
   const resolved = await resolveModelOrError(
     modelStr,
@@ -1351,6 +1398,8 @@ async function handleSingleModelChat(
       sourceFormat: sNetSourceFormat,
       endpointPath: clientRawRequest?.endpoint || "",
       requestHeaders: clientRawRequest?.headers,
+      videoTranscriptSensitive: runtimeOptions.videoTranscriptSensitive === true,
+      videoTranscriptDescriptionFingerprints,
       clientManagedResponsesContext:
         sNetSourceFormat === "openai-responses" &&
         String(clientRawRequest?.endpoint || "")
@@ -1382,6 +1431,8 @@ async function handleSingleModelChat(
               redirectCombo.config?.reasoningTransportFallback === "skip" ? "skip" : "drop",
             conversationId: runtimeOptions?.conversationId ?? null,
             managedLease: runtimeOptions.managedLease ?? null,
+            videoTranscriptSensitive: runtimeOptions.videoTranscriptSensitive === true,
+            videoTranscriptDescriptionFingerprints,
             // #7360 follow-up — see the primary handleSingleModel closure above.
             modelAbortSignal: target?.modelAbortSignal ?? null,
           },
@@ -1486,6 +1537,7 @@ async function handleSingleModelChat(
         correlationId: runtimeOptions?.correlationId ?? null,
         sessionTag: runtimeOptions?.conversationId ?? null,
         startTime: telemetry?.startTime,
+        videoTranscriptSensitive: runtimeOptions.videoTranscriptSensitive === true,
       });
     } catch {}
     return gate;
@@ -1693,7 +1745,8 @@ async function handleSingleModelChat(
           lastError,
           lastStatus,
           candidateAliases,
-          isCombo
+          isCombo,
+          runtimeOptions.videoTranscriptSensitive === true
         );
         const lastFailedConnectionId =
           excludedConnectionIds.size > 0
@@ -1853,6 +1906,8 @@ async function handleSingleModelChat(
             sessionAffinityKey: runtimeOptions.sessionAffinityKey ?? null,
             reasoningTransportFallback: runtimeOptions.reasoningTransportFallback ?? "drop",
             managedLease: runtimeOptions.managedLease ?? null,
+            videoTranscriptSensitive: runtimeOptions.videoTranscriptSensitive === true,
+            videoTranscriptDescriptionFingerprints,
           },
           runtimeOptions
         );
@@ -1888,6 +1943,7 @@ async function handleSingleModelChat(
         comboName,
         clientRawRequest,
         tlsFingerprintUsed,
+        videoTranscriptSensitive: runtimeOptions.videoTranscriptSensitive === true,
       });
 
       if (result.success) {
@@ -1990,7 +2046,13 @@ async function handleSingleModelChat(
           provider,
           model,
           providerProfile,
-          { isCombo }
+          {
+            isCombo,
+            retainedErrorText: redactVideoTranscriptSensitiveText(
+              classificationError,
+              runtimeOptions.videoTranscriptSensitive === true
+            ),
+          }
         );
 
         if (shouldFallback && !hasForcedConnection) {
@@ -2039,7 +2101,13 @@ async function handleSingleModelChat(
           provider,
           model,
           providerProfile,
-          { isCombo }
+          {
+            isCombo,
+            retainedErrorText: redactVideoTranscriptSensitiveText(
+              result.error || ANTIGRAVITY_PRE_RESPONSE_TIMEOUT_CODE,
+              runtimeOptions.videoTranscriptSensitive === true
+            ),
+          }
         );
 
         if (shouldFallback && !hasForcedConnection) {
@@ -2177,9 +2245,14 @@ async function handleSingleModelChat(
             : classify429FromError({ status: result.status, message: errorStr })
           : undefined;
       if (result.status === 429 && isDailyQuotaExhausted(errorStr)) {
-        // Parse which model is quota-limited
+        // Keep the raw provider token for routing state, but never retain it when a sensitive
-        const match = errorStr.match(/today's quota for model ([^,]+)/);
+        // upstream diagnostic could have echoed transcript text.
-        const limitedModel = match ? match[1].trim() : model;
+        const { operationalModel: limitedModel, retainedModel: retainedLimitedModel } =
+          resolveDailyQuotaModelViews(
+            errorStr,
+            model,
+            runtimeOptions.videoTranscriptSensitive === true
+          );
 
         const mlSettings = resolveModelLockoutSettings(runtimeOptions.cachedSettings);
         if (mlSettings.enabled && mlSettings.errorCodes.includes(result.status)) {
@@ -2200,7 +2273,7 @@ async function handleSingleModelChat(
             "MODEL_DAILY_QUOTA",
             JSON.stringify({
               connection: credentials.connectionId.slice(0, 8),
-              model: limitedModel,
+              model: retainedLimitedModel,
               cooldownMs: lockResult.cooldownMs,
               failureCount: lockResult.failureCount,
             })
@@ -2291,6 +2364,10 @@ async function handleSingleModelChat(
                 (failureKind === "rate_limit" || failureKind === "transient")
               ),
               isCombo,
+              retainedErrorText: redactVideoTranscriptSensitiveText(
+                errorStr,
+                runtimeOptions.videoTranscriptSensitive === true
+              ),
             }
           );
 

src/sse/handlers/chatHelpers.ts

@@ -28,6 +28,7 @@ import {
 } from "@omniroute/open-sse/utils/proxyFetch.ts";
 import { resolveProxyForConnection } from "@/lib/localDb";
 import { hasBlockingProxyAssignment } from "@/lib/db/proxies";
+import { redactVideoTranscriptSensitiveText } from "@/lib/guardrails/videoTranscriptLogRedaction";
 import {
   CircuitBreakerOpenError,
   getCircuitBreaker,
@@ -36,10 +37,10 @@ import {
 import { classify429FromError, type FailureKind } from "../../shared/utils/classify429";
 import { resolveUseUpstream429BreakerHints } from "../../shared/utils/providerHints";
 
-import { logProxyEvent } from "../../lib/proxyLogger";
-import { logTranslationEvent } from "../../lib/translatorEvents";
 import { getRuntimeProviderProfile } from "@omniroute/open-sse/services/accountFallback.ts";
 
+export { safeLogEvents } from "./chatLogEvents";
+
 // Models that explicitly cannot run on the codex/ChatGPT-Pro OAuth pool — when
 // a caller writes `codex/deepseek-v4-pro` we transparently reroute to the
 // canonical provider whose API key is configured. Saves callers from having
@@ -391,7 +392,6 @@ export function checkResourcePressureBeforeProviderWork(): ResourcePressureGuard
     return null;
   }
 }
-
 export async function executeChatWithBreaker({
   bypassCircuitBreaker,
   breaker,
@@ -425,6 +425,8 @@ export async function executeChatWithBreaker({
   reasoningTransportFallback = "drop",
   sessionAffinityKey = null,
   managedLease = null,
+  videoTranscriptSensitive = false,
+  videoTranscriptDescriptionFingerprints = [],
 }: ExecuteChatWithBreakerOptions): Promise<ExecuteChatWithBreakerResult> {
   let tlsFingerprintUsed = false;
   const normalizedTrafficType: TrafficType =
@@ -432,7 +434,6 @@ export async function executeChatWithBreaker({
       ? "shadow"
       : "production";
   const isShadowTraffic = normalizedTrafficType === "shadow";
-
   // #5217: capture the proxy actually applied during execution so the caller can
   // merge it into proxyInfo before the egress log (executors pinning a per-account
   // proxy internally otherwise leave the egress log reading "direct").
@@ -484,6 +485,8 @@ export async function executeChatWithBreaker({
             sessionAffinityKey,
             reasoningTransportFallback,
             managedLease,
+            videoTranscriptSensitive,
+            videoTranscriptDescriptionFingerprints,
             skipResourcePressureGuard: true,
             onCredentialsRefreshed: async (newCreds: any) => {
               await updateProviderCredentials(credentials.connectionId, {
@@ -537,7 +540,13 @@ export async function executeChatWithBreaker({
                 provider,
                 model,
                 providerProfile,
-                { isCombo }
+                {
+                  isCombo,
+                  retainedErrorText: redactVideoTranscriptSensitiveText(
+                    String(failure?.message || failure?.code || "stream failure"),
+                    videoTranscriptSensitive
+                  ),
+                }
               );
             },
           })
@@ -609,7 +618,7 @@ export async function executeChatWithBreaker({
 
     if (cbErr?.code === "PROXY_UNREACHABLE" || /proxy unreachable/i.test(cbErr?.message || "")) {
       const detail = cbErr?.message || "Proxy unreachable";
-      log.warn("PROXY", detail);
+      log.warn("PROXY", redactVideoTranscriptSensitiveText(detail, videoTranscriptSensitive));
       return {
         result: {
           success: false,
@@ -633,7 +642,8 @@ export function handleNoCredentials(
   lastError: string | null,
   lastStatus: number | null,
   candidateAliases?: readonly string[],
-  isCombo: boolean = false
+  isCombo: boolean = false,
+  videoTranscriptSensitive: boolean = false
 ) {
   if (credentials?.allRateLimited) {
     const errorMsg = lastError || credentials.lastError || "Unavailable";
@@ -660,7 +670,8 @@ export function handleNoCredentials(
       });
     }
 
-    log.warn("CHAT", `[${provider}/${model}] ${errorMsg} (${credentials.retryAfterHuman})`);
+    const retainedErrorMsg = redactVideoTranscriptSensitiveText(errorMsg, videoTranscriptSensitive);
+    log.warn("CHAT", `[${provider}/${model}] ${retainedErrorMsg} (${credentials.retryAfterHuman})`);
     return unavailableResponse(
       status,
       `[${provider}/${model}] ${errorMsg}`,
@@ -845,82 +856,6 @@ export function applyExecutorProxyToInfo(
   };
 }
 
-// Async because the egress-IP lookup lazy-imports proxyEgress; callers treat
-// this as fire-and-forget logging (the internal try/catch swallows everything).
-export async function safeLogEvents({
-  result,
-  proxyInfo,
-  proxyLatency,
-  provider,
-  model,
-  sourceFormat,
-  targetFormat,
-  credentials,
-  comboName,
-  clientRawRequest,
-  tlsFingerprintUsed = false,
-}) {
-  try {
-    const rawIp =
-      clientRawRequest?.headers?.["x-forwarded-for"] ||
-      clientRawRequest?.headers?.["x-real-ip"] ||
-      clientRawRequest?.headers?.["cf-connecting-ip"] ||
-      null;
-    const rawIpValue = Array.isArray(rawIp) ? rawIp[0] : rawIp;
-    const clientIp = typeof rawIpValue === "string" ? rawIpValue.split(",")[0].trim() : null;
-
-    // Resolve the egress IP (the IP the upstream actually saw) from cache — never
-    // blocking the request. Warm it in the background for next time. null until
-    // the first warm completes; direct (no proxy) is also tracked.
-    let egressIp: string | null = null;
-    try {
-      const { getCachedEgressIp, warmEgressIp } = await import("../../lib/proxyEgress");
-      const { proxyConfigToUrl } = await import("@omniroute/open-sse/utils/proxyDispatcher.ts");
-      const proxyUrl = proxyInfo?.proxy ? proxyConfigToUrl(proxyInfo.proxy) : null;
-      egressIp = getCachedEgressIp(proxyUrl);
-      warmEgressIp(proxyUrl);
-    } catch {
-      // egress visibility is best-effort; never break the request path
-    }
-
-    logProxyEvent({
-      status: result.success
-        ? "success"
-        : result.status === 408 || result.status === 504
-          ? "timeout"
-          : "error",
-      proxy: proxyInfo?.proxy || null,
-      level: proxyInfo?.level || "direct",
-      levelId: proxyInfo?.levelId || null,
-      provider,
-      targetUrl: `${provider}/${model}`,
-      clientIp,
-      egressIp,
-      latencyMs: proxyLatency,
-      error: result.success ? null : result.error || null,
-      connectionId: credentials.connectionId,
-      comboId: comboName || null,
-      account: credentials.connectionId?.slice(0, 8) || null,
-      tlsFingerprint: tlsFingerprintUsed,
-    });
-  } catch {}
-
-  try {
-    logTranslationEvent({
-      provider,
-      model,
-      sourceFormat,
-      targetFormat,
-      status: result.success ? "success" : "error",
-      statusCode: result.success ? 200 : result.status || 500,
-      latency: proxyLatency,
-      endpoint: clientRawRequest?.endpoint || "/v1/chat/completions",
-      connectionId: credentials.connectionId || null,
-      comboName: comboName || null,
-    });
-  } catch {}
-}
-
 export function withSessionHeader(response: Response, sessionId: string | null): Response {
   if (!response || !sessionId) return response;
 

src/sse/handlers/chatLogEvents.ts

@@ -0,0 +1,107 @@
+import { redactVideoTranscriptSensitiveText } from "@/lib/guardrails/videoTranscriptLogRedaction";
+import { logProxyEvent } from "../../lib/proxyLogger";
+import { logTranslationEvent } from "../../lib/translatorEvents";
+
+type ProxyLogConfig = {
+  host: string;
+  port: number | string;
+  type: string;
+};
+
+export interface SafeLogEventsInput {
+  clientRawRequest?: {
+    endpoint?: string;
+    headers?: Record<string, string | string[] | undefined>;
+  } | null;
+  comboName?: string | null;
+  credentials: { connectionId?: string | null };
+  model: string;
+  provider: string;
+  proxyInfo?: {
+    level?: string;
+    levelId?: string | null;
+    proxy?: ProxyLogConfig | null;
+  } | null;
+  proxyLatency: number;
+  result: { error?: unknown; status?: number; success: boolean };
+  sourceFormat: string;
+  targetFormat: string;
+  tlsFingerprintUsed?: boolean;
+  videoTranscriptSensitive?: boolean;
+}
+
+/** Retain safe proxy/translation metadata without retaining a sensitive provider error echo. */
+export async function safeLogEvents({
+  result,
+  proxyInfo,
+  proxyLatency,
+  provider,
+  model,
+  sourceFormat,
+  targetFormat,
+  credentials,
+  comboName,
+  clientRawRequest,
+  tlsFingerprintUsed = false,
+  videoTranscriptSensitive = false,
+}: SafeLogEventsInput): Promise<void> {
+  try {
+    const rawIp =
+      clientRawRequest?.headers?.["x-forwarded-for"] ||
+      clientRawRequest?.headers?.["x-real-ip"] ||
+      clientRawRequest?.headers?.["cf-connecting-ip"] ||
+      null;
+    const rawIpValue = Array.isArray(rawIp) ? rawIp[0] : rawIp;
+    const clientIp = typeof rawIpValue === "string" ? rawIpValue.split(",")[0].trim() : null;
+
+    let egressIp: string | null = null;
+    try {
+      const { getCachedEgressIp, warmEgressIp } = await import("../../lib/proxyEgress");
+      const { proxyConfigToUrl } = await import("@omniroute/open-sse/utils/proxyDispatcher.ts");
+      const proxyUrl = proxyInfo?.proxy ? proxyConfigToUrl(proxyInfo.proxy) : null;
+      egressIp = getCachedEgressIp(proxyUrl);
+      warmEgressIp(proxyUrl);
+    } catch {
+      // Egress visibility is best-effort and never breaks the request path.
+    }
+
+    logProxyEvent({
+      account: credentials.connectionId?.slice(0, 8) || null,
+      clientIp,
+      comboId: comboName || null,
+      connectionId: credentials.connectionId,
+      egressIp,
+      error:
+        result.success || !result.error
+          ? null
+          : redactVideoTranscriptSensitiveText(String(result.error), videoTranscriptSensitive),
+      latencyMs: proxyLatency,
+      level: proxyInfo?.level || "direct",
+      levelId: proxyInfo?.levelId || null,
+      provider,
+      proxy: proxyInfo?.proxy || null,
+      status: result.success
+        ? "success"
+        : result.status === 408 || result.status === 504
+          ? "timeout"
+          : "error",
+      targetUrl: `${provider}/${model}`,
+      tlsFingerprint: tlsFingerprintUsed,
+    });
+  } catch {}
+
+  try {
+    logTranslationEvent({
+      comboName: comboName || null,
+      connectionId: credentials.connectionId || null,
+      endpoint: clientRawRequest?.endpoint || "/v1/chat/completions",
+      latency: proxyLatency,
+      model,
+      provider,
+      sourceFormat,
+      status: result.success ? "success" : "error",
+      statusCode: result.success ? 200 : result.status || 500,
+      targetFormat,
+    });
+  } catch {}
+}

src/sse/handlers/chatPredicates.ts

@@ -1,6 +1,7 @@
+import { getTrustedLocalRateLimitResponse } from "@omniroute/open-sse/services/rateLimitManager/errors";
+
 import { isLocalStreamLifecycleError } from "../../shared/utils/circuitBreaker";
 import { isRequestScopedUpstreamFailure } from "./comboFailureLogging";
-import { getTrustedLocalRateLimitResponse } from "@omniroute/open-sse/services/rateLimitManager/errors";
 
 export const PROVIDER_BREAKER_FAILURE_STATUSES = new Set([408, 500, 502, 503, 504]);
 
@@ -69,3 +70,32 @@ export function resolveStreamReadinessClassificationError(
   }
   return fallback;
 }
+
+export interface DailyQuotaModelViews {
+  /** Model key used by model-lockout state and future routing decisions. */
+  operationalModel: string;
+  /** Safe representation allowed in retained application logs. */
+  retainedModel: string;
+}
+
+/**
+ * Ordinary requests preserve the exact upstream model token used by existing quota routing.
+ * For transcript-sensitive requests, the provider diagnostic is untrusted request-derived text:
+ * keep classifying the raw error at the caller, but never promote its token into retained lockout
+ * state. The already-resolved request model is the server-owned key and safe retained view.
+ */
+export function resolveDailyQuotaModelViews(
+  errorText: string,
+  resolvedModel: string,
+  videoTranscriptSensitive: boolean
+): DailyQuotaModelViews {
+  const match = errorText.match(/today's quota for model ([^,]+)/);
+  const upstreamModel = match?.[1]?.trim();
+  const operationalModel = videoTranscriptSensitive
+    ? resolvedModel
+    : upstreamModel || resolvedModel;
+  return {
+    operationalModel,
+    retainedModel: operationalModel,
+  };
+}

src/sse/handlers/rejectedRequestUsage.ts

@@ -18,6 +18,11 @@
  * never turn into a second failure on the response path.
  */
 import { saveCallLog, saveRequestUsage } from "@/lib/usageDb";
+import {
+  containsVideoTranscriptForLog,
+  omitVideoTranscriptForLog,
+  VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+} from "@/lib/guardrails/videoTranscriptLogRedaction";
 
 export interface RejectedRequestUsageInput {
   status: number;
@@ -44,6 +49,8 @@ export interface RejectedRequestUsageInput {
    * dashboard log detail had no request to inspect — see #7360 follow-up.
    */
   requestBody?: unknown;
+  /** Trusted request-scoped signal; raw text remains available to routing before this boundary. */
+  videoTranscriptSensitive?: boolean;
 }
 
 export async function recordRejectedRequestUsage(input: RejectedRequestUsageInput): Promise<void> {
@@ -64,10 +71,20 @@ export async function recordRejectedRequestUsage(input: RejectedRequestUsageInpu
     connectionId = undefined,
     startTime,
     requestBody = null,
+    videoTranscriptSensitive = false,
   } = input;
 
   const now = Date.now();
   const duration = typeof startTime === "number" ? now - startTime : 0;
+  // This sink owns a request body and must defend itself if an upstream caller forgets the
+  // request-scoped bit. The explicit bit is still required after a guardrail replaced the raw
+  // carrier with generated text, which is why both sources participate.
+  const transcriptSensitive =
+    videoTranscriptSensitive || containsVideoTranscriptForLog(requestBody);
+  const retainedError = transcriptSensitive ? VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER : error || null;
+  const retainedRequestBody = transcriptSensitive
+    ? omitVideoTranscriptForLog(requestBody)
+    : requestBody;
 
   // 1. call_logs — preserves /dashboard/logs visibility (unchanged behavior).
   await saveCallLog({
@@ -81,8 +98,8 @@ export async function recordRejectedRequestUsage(input: RejectedRequestUsageInpu
     connectionId,
     duration,
     tokens: {},
-    error: error || null,
+    error: retainedError,
-    requestBody,
+    requestBody: retainedRequestBody,
     comboName,
     comboStepId,
     comboExecutionKey,

src/sse/services/auth.ts

@@ -2607,10 +2607,11 @@ export async function markAccountUnavailable(
   providerProfile = null,
   options: {
     persistUnavailableState?: boolean;
-    /** Caller is the combo engine — it records its own model-level lockouts. */
     isCombo?: boolean;
+    retainedErrorText?: string;
   } = {}
 ) {
+  const retainedErrorText = options.retainedErrorText ?? errorText;
   const currentMutex = markMutexes.get(connectionId) || Promise.resolve();
   let resolveMutex: (() => void) | undefined;
   markMutexes.set(
@@ -2619,10 +2620,8 @@ export async function markAccountUnavailable(
       resolveMutex = resolve;
     })
   );
-
   try {
     await currentMutex;
-
     // STRICT_ZERO_COST: this connection just failed (whatever the reason) —
     // drop any cached "SAFE" free-allowance reading for it immediately rather
     // than waiting out the TTL, so the very next candidate-pool build reads a
@@ -2736,14 +2735,13 @@ export async function markAccountUnavailable(
     // the opt-in setting probeCanDisable restores the historical behavior.
     if (await shouldIsolateProbeFailures()) {
       await updateProviderConnection(connectionId, {
-        // lastError kept RAW (full text) — maximal probe visibility; the
+        // Probe visibility keeps full retained text; sensitive callers supply an omission marker.
-        // divergence vs the normal path's slice(0,100) is intentional.
         // backoffLevel is deliberately NOT written: a positive backoff
         // triggers the selection-time auto-decay (resetConnectionBackoff,
         // auth.ts getProviderCredentials) which wipes lastError back to
         // NULL on the next attempt — silently destroying the probe record.
         // The backoff is also routing state a probe must not touch (#9817).
-        lastError: errorText,
+        lastError: retainedErrorText,
         lastErrorType: fallbackResult.reason || null,
         errorCode: status,
         lastErrorAt: new Date().toISOString(),
@@ -3148,7 +3146,7 @@ export async function markAccountUnavailable(
       return { shouldFallback: true, cooldownMs: lockout.cooldownMs };
     }
 
-    const errorMsg = describeUpstreamFailure(errorText);
+    const errorMsg = describeUpstreamFailure(retainedErrorText);
 
     // T09: Codex per-scope lockout (do not block the whole account globally).
     if (

tests/unit/chat-daily-quota-model-retention.test.ts

@@ -0,0 +1,42 @@
+import assert from "node:assert/strict";
+import test from "node:test";
+
+import { resolveDailyQuotaModelViews } from "../../src/sse/handlers/chatPredicates.ts";
+
+test("daily quota model uses the upstream model token for ordinary operational and retained views", () => {
+  assert.deepEqual(
+    resolveDailyQuotaModelViews(
+      "You have exceeded today's quota for model moonshotai/Kimi-K2.5, try tomorrow",
+      "Kimi-K2.5",
+      false
+    ),
+    {
+      operationalModel: "moonshotai/Kimi-K2.5",
+      retainedModel: "moonshotai/Kimi-K2.5",
+    }
+  );
+});
+
+test("daily quota uses the server-resolved key when the upstream token may echo a transcript", () => {
+  const transcriptSentinel = "PRIVATE_DAILY_QUOTA_TRANSCRIPT_SENTINEL";
+  const views = resolveDailyQuotaModelViews(
+    `You have exceeded today's quota for model ${transcriptSentinel}, try tomorrow`,
+    "server-resolved-model",
+    true
+  );
+
+  assert.equal(views.operationalModel, "server-resolved-model");
+  assert.equal(views.retainedModel, "server-resolved-model");
+  assert.equal(views.operationalModel.includes(transcriptSentinel), false);
+  assert.equal(views.retainedModel.includes(transcriptSentinel), false);
+});
+
+test("daily quota model falls back to the server-resolved model when no token is present", () => {
+  assert.deepEqual(
+    resolveDailyQuotaModelViews("daily quota exhausted", "server-resolved-model", true),
+    {
+      operationalModel: "server-resolved-model",
+      retainedModel: "server-resolved-model",
+    }
+  );
+});

tests/unit/chat-helpers.test.ts

@@ -6,7 +6,11 @@ import path from "node:path";
 import net from "node:net";
 
 const TEST_DATA_DIR = fs.mkdtempSync(path.join(os.tmpdir(), "omniroute-chat-helpers-"));
+const TEST_APP_LOG_DIR = fs.mkdtempSync(path.join(os.tmpdir(), "omniroute-chat-helpers-log-"));
+const TEST_APP_LOG_PATH = path.join(TEST_APP_LOG_DIR, "app.log");
 process.env.DATA_DIR = TEST_DATA_DIR;
+process.env.APP_LOG_TO_FILE = "true";
+process.env.APP_LOG_FILE_PATH = TEST_APP_LOG_PATH;
 
 const core = await import("../../src/lib/db/core.ts");
 const providersDb = await import("../../src/lib/db/providers.ts");
@@ -22,6 +26,8 @@ const {
 } = await import("../../src/sse/handlers/chatHelpers.ts");
 const { getCircuitBreaker, resetAllCircuitBreakers, STATE } =
   await import("../../src/shared/utils/circuitBreaker.ts");
+const proxyLogger = await import("../../src/lib/proxyLogger.ts");
+const { logger: appLogger } = await import("../../src/shared/utils/logger.ts");
 // DATA_DIR must be fixed before these modules load; keep this test seam dynamic.
 const { setTlsClientForTest } = await import("../../open-sse/utils/proxyFetch.ts");
 
@@ -32,6 +38,31 @@ async function resetStorage() {
   fs.mkdirSync(TEST_DATA_DIR, { recursive: true });
 }
 
+async function readAppLogWhen(
+  predicate: (contents: string) => boolean,
+  timeoutMs = 4_000
+): Promise<string> {
+  const startedAt = Date.now();
+  while (Date.now() - startedAt < timeoutMs) {
+    const contents = fs.existsSync(TEST_APP_LOG_PATH)
+      ? fs.readFileSync(TEST_APP_LOG_PATH, "utf8")
+      : "";
+    if (predicate(contents)) return contents;
+    await new Promise((resolve) => setTimeout(resolve, 50));
+  }
+  return fs.existsSync(TEST_APP_LOG_PATH) ? fs.readFileSync(TEST_APP_LOG_PATH, "utf8") : "";
+}
+
+async function flushAppLogger(): Promise<void> {
+  await new Promise<void>((resolveFlush) => {
+    try {
+      appLogger.flush(() => resolveFlush());
+    } catch {
+      resolveFlush();
+    }
+  });
+}
+
 async function seedConnection(provider, overrides = {}) {
   return providersDb.createProviderConnection({
     provider,
@@ -420,6 +451,42 @@ test("handleNoCredentials returns Retry-After when every account is rate limited
   assert.match(json.error.message, /\[openai\/gpt-4o-mini\] Quota exceeded/);
 });
 
+test("handleNoCredentials omits transcript-sensitive upstream text from application logs", async () => {
+  const rawCue = "PRIVATE_NO_CREDENTIALS_VIDEO_TRANSCRIPT_SENTINEL";
+  const provider = `fu05-${Math.random().toString(16).slice(2)}`;
+
+  const response = handleNoCredentials(
+    {
+      allRateLimited: true,
+      retryAfter: new Date(Date.now() + 30_000).toISOString(),
+      retryAfterHuman: "reset after 30s",
+      lastErrorCode: 429,
+    },
+    "conn_video_transcript",
+    provider,
+    "fixture-model",
+    rawCue,
+    429,
+    undefined,
+    false,
+    true
+  );
+
+  const body = (await response.json()) as { error?: { message?: string } };
+  assert.match(body.error?.message ?? "", new RegExp(rawCue));
+
+  await flushAppLogger();
+  const contents = await readAppLogWhen(
+    (value) => value.includes(provider) && value.includes("omitted: video transcript")
+  );
+  const matchingLines = contents
+    .split("\n")
+    .filter((line) => line.includes(provider))
+    .join("\n");
+  assert.doesNotMatch(matchingLines, new RegExp(rawCue));
+  assert.match(matchingLines, /omitted: video transcript/);
+});
+
 test("handleNoCredentials returns structured model_cooldown when every credential for the model is cooling down", async () => {
   const retryAfter = new Date(Date.now() + 12_000).toISOString();
   const response = handleNoCredentials(
@@ -553,6 +620,60 @@ test("executeChatWithBreaker converts proxy fast-fail errors", async () => {
   }
 });
 
+test("executeChatWithBreaker omits transcript echoes from proxy fast-fail logs", async () => {
+  const rawCue = "PRIVATE_PROXY_FAST_FAIL_VIDEO_TRANSCRIPT_SENTINEL";
+  const error = new Error(`Proxy unreachable: ${rawCue}`);
+  (error as Error & { code?: string }).code = "PROXY_UNREACHABLE";
+  const rejectingBreaker = {
+    execute: async () => {
+      throw error;
+    },
+  };
+
+  await flushAppLogger();
+  const before = fs.existsSync(TEST_APP_LOG_PATH)
+    ? fs.readFileSync(TEST_APP_LOG_PATH, "utf8").length
+    : 0;
+
+  const credentials = {
+    connectionId: "conn_sensitive_proxy",
+    apiKey: "sk-openai-helper",
+    providerSpecificData: {},
+  };
+  const proxyResult = await executeChatWithBreaker({
+    bypassCircuitBreaker: false,
+    breaker: rejectingBreaker,
+    body: { model: "openai/gpt-4o-mini" },
+    provider: "openai",
+    model: "gpt-4o-mini",
+    refreshedCredentials: credentials,
+    proxyInfo: null,
+    log: console,
+    clientRawRequest: null,
+    credentials,
+    apiKeyInfo: null,
+    userAgent: "",
+    comboName: null,
+    comboStrategy: null,
+    isCombo: false,
+    extendedContext: false,
+    comboStepId: null,
+    comboExecutionKey: null,
+    videoTranscriptSensitive: true,
+  });
+
+  assert.equal(proxyResult.result.status, 503);
+  assert.match(String(proxyResult.result.error || ""), new RegExp(rawCue));
+
+  await flushAppLogger();
+  const contents = await readAppLogWhen(
+    (value) => value.slice(before).includes("omitted: video transcript") || value.includes(rawCue)
+  );
+  const retainedLines = contents.slice(before);
+  assert.doesNotMatch(retainedLines, new RegExp(rawCue));
+  assert.match(retainedLines, /omitted: video transcript/);
+});
+
 test("executeChatWithBreaker preserves account TLS scope when a proxy bypasses to direct", async () => {
   const server = net.createServer((socket) => socket.end());
   const listening = Promise.withResolvers<void>();
@@ -674,6 +795,49 @@ test("safeLogEvents tolerates success and timeout payloads", () => {
   });
 });
 
+test("safeLogEvents preserves a null error for failures without error details", async () => {
+  const provider = "fu05-missing-error";
+  await safeLogEvents({
+    result: { success: false, status: 502 },
+    proxyInfo: { proxy: null, level: "direct", levelId: null },
+    proxyLatency: 25,
+    provider,
+    model: "fixture-model",
+    sourceFormat: "openai-chat",
+    targetFormat: "openai-chat",
+    credentials: { connectionId: "conn-missing-error" },
+    comboName: null,
+    clientRawRequest: { endpoint: "/v1/chat/completions" },
+  });
+
+  const entry = proxyLogger.getProxyLogs({ provider, limit: 1 })[0];
+  assert.ok(entry);
+  assert.equal(entry.error, null);
+});
+
+test("safeLogEvents omits failure text for transcript-sensitive requests", async () => {
+  const rawCue = "PRIVATE_PROXY_LOG_VIDEO_TRANSCRIPT_SENTINEL";
+  const provider = "fu05-transcript-sensitive";
+  await safeLogEvents({
+    result: { success: false, status: 502, error: rawCue },
+    proxyInfo: { proxy: null, level: "direct", levelId: null },
+    proxyLatency: 25,
+    provider,
+    model: "fixture-model",
+    sourceFormat: "openai-chat",
+    targetFormat: "openai-chat",
+    credentials: { connectionId: "conn-video-transcript" },
+    comboName: null,
+    clientRawRequest: { endpoint: "/v1/chat/completions" },
+    videoTranscriptSensitive: true,
+  });
+
+  const entry = proxyLogger.getProxyLogs({ provider, limit: 1 })[0];
+  assert.ok(entry);
+  assert.equal(String(entry.error).includes(rawCue), false);
+  assert.match(String(entry.error), /omitted: video transcript/);
+});
+
 test("withSessionHeader adds headers to mutable and immutable responses", async () => {
   const mutable = withSessionHeader(new Response("ok"), "sess_mutable");
   const immutable = withSessionHeader(Response.redirect("https://example.com"), "sess_redirect");

tests/unit/chatcore-attempt-logging.test.ts

@@ -14,6 +14,9 @@ const testDataDir = fs.mkdtempSync(path.join(os.tmpdir(), "omni-attempt-logging-
 process.env.DATA_DIR = testDataDir;
 
 const coreDb = await import("../../src/lib/db/core.ts");
+const eventBus = await import("../../src/lib/events/eventBus.ts");
+const { fingerprintVideoTranscriptDescription } =
+  await import("../../src/lib/guardrails/videoTranscriptLogRedaction.ts");
 const { getCallLogById } = await import("../../src/lib/usage/callLogs.ts");
 const { persistAttemptLogs } = await import("../../open-sse/handlers/chatCore/attemptLogging.ts");
 
@@ -48,6 +51,7 @@ function baseCtx(overrides: Record<string, unknown> = {}) {
     tokensCompressed: 0,
     apiKeyInfo: { id: "key-1", name: "Key One" },
     noLogEnabled: false,
+    videoTranscriptSensitive: false,
     ...overrides,
   } as Parameters<typeof persistAttemptLogs>[1];
 }
@@ -136,3 +140,75 @@ test("connectionId falls back to credentials.connectionId when null, and error i
   assert.equal(row.status, 502);
   assert.match(String(row.error ?? ""), /upstream boom/);
 });
+
+test("omits non-stream response echoes for a transcript-sensitive attempt", async () => {
+  const id = "attempt-video-transcript-sensitive-1";
+  const traceId = "trace-video-transcript-sensitive-1";
+  const rawCue = "private attempt-log subtitle echo sentinel";
+  const description = `[Video description: transcript[source=embedded] text=${JSON.stringify(rawCue)}]`;
+  persistAttemptLogs(
+    {
+      status: 502,
+      error: `upstream echoed ${rawCue}`,
+      responseBody: { choices: [{ message: { content: rawCue } }] },
+      providerResponse: {
+        choices: [{ message: { content: rawCue } }],
+        warning: `safety filter echoed ${rawCue}`,
+      },
+      clientResponse: { choices: [{ message: { content: rawCue } }] },
+    },
+    baseCtx({
+      detailedLoggingEnabled: true,
+      pendingRequestId: id,
+      skillRequestId: "skill-video-transcript-sensitive",
+      traceId,
+      body: {
+        messages: [
+          {
+            role: "user",
+            content: description,
+          },
+        ],
+      },
+      reqLogger: {
+        getPipelinePayloads: () => ({}),
+        isVideoTranscriptSensitive: () => true,
+      },
+      videoTranscriptDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+      videoTranscriptSensitive: true,
+    })
+  );
+
+  const row = await pollForCallLog(id);
+  assert.ok(row);
+  const serialized = JSON.stringify({
+    error: row.error,
+    pipelinePayloads: row.pipelinePayloads,
+    responseBody: row.responseBody,
+  });
+  assert.equal(serialized.includes(rawCue), false);
+  assert.match(serialized, /omitted: video transcript/);
+
+  await new Promise<void>((resolve) => setImmediate(resolve));
+  const lifecycle = eventBus
+    .getEventHistory(undefined, 100)
+    .find(
+      (entry) =>
+        entry.event === "request.failed" &&
+        (entry.payload as { id?: unknown } | undefined)?.id === traceId
+    );
+  assert.ok(lifecycle, "request.failed must be retained for dashboard lifecycle cleanup");
+  const lifecyclePayload = JSON.stringify(lifecycle.payload);
+  assert.equal(lifecyclePayload.includes(rawCue), false);
+  assert.match(lifecyclePayload, /omitted: video transcript/);
+
+  const auditRow = coreDb
+    .getDbInstance()
+    .prepare(
+      "SELECT details FROM audit_log WHERE action = 'provider.warning' AND request_id = ? ORDER BY id DESC LIMIT 1"
+    )
+    .get("skill-video-transcript-sensitive") as { details?: string } | undefined;
+  assert.ok(auditRow, "provider.warning existence must survive transcript redaction");
+  assert.equal(String(auditRow.details).includes(rawCue), false);
+  assert.match(String(auditRow.details), /omitted: video transcript/);
+});

tests/unit/chatcore-executor-proxy.test.ts

@@ -17,13 +17,11 @@ process.env.DATA_DIR = testDataDir;
 // Dynamic imports AFTER DATA_DIR is set so core.ts picks up the temp path.
 const coreDb = await import("../../src/lib/db/core.ts");
 const upstreamProxyDb = await import("../../src/lib/db/upstreamProxy.ts");
-const { resolveExecutorWithProxy } = await import(
+const { resolveExecutorWithProxy } =
-  "../../open-sse/handlers/chatCore/executorProxy.ts"
+  await import("../../open-sse/handlers/chatCore/executorProxy.ts");
-);
 const { getExecutor } = await import("../../open-sse/executors/index.ts");
-const { clearUpstreamProxyConfigCache } = await import(
+const { clearUpstreamProxyConfigCache } =
-  "../../open-sse/handlers/chatCore/comboContextCache.ts"
+  await import("../../open-sse/handlers/chatCore/comboContextCache.ts");
-);
 
 before(async () => {
   await coreDb.ensureDbInitialized();
@@ -137,3 +135,68 @@ test("connection override wins over provider mode 'fallback'", async () => {
   // Connection override short-circuits to the passthrough executor, not the fallback wrapper.
   assert.equal(exec, await getExecutor("cliproxyapi"));
 });
+
+test("fallback diagnostics omit transcript echoes while preserving operational errors", async () => {
+  const nativeSentinel = "PRIVATE_NATIVE_VIDEO_TRANSCRIPT_SENTINEL";
+  const thrownFallbackSentinel = "PRIVATE_THROWN_FALLBACK_VIDEO_TRANSCRIPT_SENTINEL";
+  const statusFallbackSentinel = "PRIVATE_STATUS_FALLBACK_VIDEO_TRANSCRIPT_SENTINEL";
+  const retainedLogs: string[] = [];
+  const log = {
+    info: (...args: unknown[]) => retainedLogs.push(args.map(String).join(" ")),
+    error: (...args: unknown[]) => retainedLogs.push(args.map(String).join(" ")),
+  };
+
+  await upstreamProxyDb.upsertUpstreamProxyConfig({
+    providerId: "openai",
+    mode: "fallback",
+    enabled: true,
+  });
+  clearUpstreamProxyConfigCache("openai");
+
+  const nativeExecutor = await getExecutor("openai");
+  const fallbackExecutor = await getExecutor("cliproxyapi");
+  const originalNativeExecute = nativeExecutor.execute;
+  const originalFallbackExecute = fallbackExecutor.execute;
+  const input = {
+    model: "video-model",
+    body: { messages: [{ role: "user", content: "describe the video" }] },
+    stream: false,
+    credentials: { apiKey: "test-key" },
+    videoTranscriptSensitive: true,
+  };
+
+  try {
+    nativeExecutor.execute = async () => {
+      throw new Error(nativeSentinel);
+    };
+    fallbackExecutor.execute = async () => {
+      throw new Error(thrownFallbackSentinel);
+    };
+
+    const thrownWrapper = await resolveExecutorWithProxy("openai", log);
+    await assert.rejects(thrownWrapper.execute(input), new RegExp(thrownFallbackSentinel));
+
+    nativeExecutor.execute = async () => ({
+      response: new Response("retryable", { status: 500 }),
+      url: "https://native.example.test/v1/chat/completions",
+      headers: {},
+      transformedBody: input.body,
+    });
+    fallbackExecutor.execute = async () => {
+      throw new Error(statusFallbackSentinel);
+    };
+
+    const statusWrapper = await resolveExecutorWithProxy("openai", log);
+    await assert.rejects(statusWrapper.execute(input), new RegExp(statusFallbackSentinel));
+
+    const retained = retainedLogs.join("\n");
+    for (const sentinel of [nativeSentinel, thrownFallbackSentinel, statusFallbackSentinel]) {
+      assert.doesNotMatch(retained, new RegExp(sentinel));
+    }
+    assert.match(retained, /omitted: video transcript/);
+  } finally {
+    nativeExecutor.execute = originalNativeExecute;
+    fallbackExecutor.execute = originalFallbackExecute;
+    clearUpstreamProxyConfigCache("openai");
+  }
+});

tests/unit/chatcore-extracted-modules-3821.test.ts

@@ -38,7 +38,11 @@ test("sanitizeChatRequestBody: Responses target maps max_tokens → max_output_t
   // #9161: token-field selection keys on the OUTBOUND (target) protocol only — a
   // Responses-shaped SOURCE no longer forces max_output_tokens (see
   // codex-responses-to-chat-9161.test.ts for that direction).
-  const out = sanitizeChatRequestBody({ max_tokens: 128 }, FORMATS.OPENAI, FORMATS.OPENAI_RESPONSES);
+  const out = sanitizeChatRequestBody(
+    { max_tokens: 128 },
+    FORMATS.OPENAI,
+    FORMATS.OPENAI_RESPONSES
+  );
   assert.equal(out.max_output_tokens, 128);
   assert.equal(out.max_tokens, undefined);
 });
@@ -80,6 +84,7 @@ test("checkIdempotencyCache returns { hit:null, idempotencyKey } on a miss", asy
     effectiveServiceTier: undefined,
     startTime: 0,
     log: undefined,
+    videoTranscriptSensitive: false,
   });
   assert.equal(result.hit, null);
   // #6558: the raw header key is now namespaced by provider/model + a messages
@@ -115,13 +120,42 @@ test("checkIdempotencyCache returns a hit Response reusing the same key after a
     effectiveServiceTier: undefined,
     startTime: 0,
     log: undefined,
+    videoTranscriptSensitive: false,
   });
 
-  assert.equal(result.idempotencyKey, key, "the resolved key is returned for the save site to reuse");
+  assert.equal(
+    result.idempotencyKey,
+    key,
+    "the resolved key is returned for the save site to reuse"
+  );
   assert.ok(result.hit, "a cached entry produces a hit");
   assert.equal(result.hit!.response.headers.get("X-OmniRoute-Idempotent"), "true");
 });
 
+test("checkIdempotencyCache bypasses an existing transcript-sensitive entry", async () => {
+  const rawKey = "idem-private-video-3821";
+  const key = composeIdempotencyKey({
+    rawKey,
+    provider: "openai",
+    model: "gpt-4.1",
+    messages: undefined,
+  })!;
+  saveIdempotency(key, { content: "PRIVATE_IDEMPOTENCY_TRANSCRIPT_SENTINEL" }, 200);
+
+  const result = await checkIdempotencyCache({
+    clientRawRequest: { headers: new Headers({ "idempotency-key": rawKey }) },
+    provider: "openai",
+    model: "gpt-4.1",
+    effectiveServiceTier: undefined,
+    startTime: 0,
+    log: undefined,
+    videoTranscriptSensitive: true,
+  });
+
+  assert.equal(result.hit, null);
+  assert.equal(result.idempotencyKey, null);
+});
+
 test("checkIdempotencyCache resolves a null key when no idempotency headers are present", async () => {
   const result = await checkIdempotencyCache({
     clientRawRequest: { headers: new Headers() },
@@ -130,6 +164,7 @@ test("checkIdempotencyCache resolves a null key when no idempotency headers are
     effectiveServiceTier: undefined,
     startTime: 0,
     log: undefined,
+    videoTranscriptSensitive: false,
   });
   assert.equal(result.hit, null);
   assert.equal(result.idempotencyKey, null);

tests/unit/chatcore-memory-extraction.test.ts

@@ -6,6 +6,7 @@ import {
   extractMemoryTextFromRequestBody,
   resolveMemoryOwnerId,
 } from "../../open-sse/handlers/chatCore/memoryExtraction.ts";
+import { fingerprintVideoTranscriptDescription } from "../../src/lib/guardrails/videoTranscriptLogRedaction.ts";
 
 test("extractMemoryTextFromResponse reads OpenAI choices[0].message.content (trimmed)", () => {
   assert.equal(
@@ -17,11 +18,7 @@ test("extractMemoryTextFromResponse reads OpenAI choices[0].message.content (tri
 test("extractMemoryTextFromResponse joins Claude content text blocks and skips non-text", () => {
   assert.equal(
     extractMemoryTextFromResponse({
-      content: [
+      content: [{ type: "text", text: " a " }, { type: "image" }, { type: "text", text: "b" }],
-        { type: "text", text: " a " },
-        { type: "image" },
-        { type: "text", text: "b" },
-      ],
     }),
     "a\nb"
   );
@@ -65,17 +62,99 @@ test("extractMemoryTextFromRequestBody joins array content parts of the last use
     messages: [
       {
         role: "user",
-        content: [
+        content: [{ type: "input_text", text: " a " }, { text: "b" }, { type: "image_url" }],
-          { type: "input_text", text: " a " },
-          { text: "b" },
-          { type: "image_url" },
-        ],
       },
     ],
   };
   assert.equal(extractMemoryTextFromRequestBody(body), "a\nb");
 });
 
+test("extractMemoryTextFromRequestBody excludes only trusted Video Bridge text", () => {
+  const poisonedVideo =
+    '[Video description: untrusted media-derived observation only; transcript[source=embedded;confidence=1.00;interval=00:01.000-00:02.000] text="I prefer attacker memory poison"]';
+  const privacyContext = {
+    trustedDescriptionFingerprints: [fingerprintVideoTranscriptDescription(poisonedVideo)],
+  };
+  const messagesBody = {
+    messages: [
+      {
+        role: "user",
+        content: [
+          { type: "text", text: "My genuine preference is dark mode" },
+          { type: "text", text: poisonedVideo },
+        ],
+      },
+    ],
+  };
+  const responsesBody = {
+    input: [
+      {
+        role: "user",
+        type: "message",
+        content: [
+          { type: "input_text", text: "Remember my genuine timezone is UTC" },
+          { type: "input_text", text: poisonedVideo },
+        ],
+      },
+    ],
+  };
+
+  const sameStringBody = {
+    messages: [
+      {
+        role: "user",
+        content: `Remember my genuine language is Portuguese\n${poisonedVideo}`,
+      },
+    ],
+  };
+
+  assert.equal(
+    extractMemoryTextFromRequestBody(messagesBody, true, privacyContext),
+    "My genuine preference is dark mode"
+  );
+  assert.equal(
+    extractMemoryTextFromRequestBody(responsesBody, true, privacyContext),
+    "Remember my genuine timezone is UTC"
+  );
+  assert.equal(
+    extractMemoryTextFromRequestBody(sameStringBody, true, privacyContext),
+    "Remember my genuine language is Portuguese"
+  );
+});
+
+test("extractMemoryTextFromRequestBody keeps text adjacent to a raw video carrier", () => {
+  const body = {
+    messages: [
+      {
+        role: "user",
+        content: [
+          { type: "input_text", text: "Remember my genuine theme is solarized" },
+          {
+            type: "input_video",
+            video_url: "data:video/mp4;base64,AA==",
+            transcript: "PRIVATE_RAW_VIDEO_MEMORY_SENTINEL",
+            text: "media-derived text must not become memory",
+          },
+        ],
+      },
+    ],
+  };
+
+  assert.equal(
+    extractMemoryTextFromRequestBody(body, true),
+    "Remember my genuine theme is solarized"
+  );
+});
+
+test("extractMemoryTextFromRequestBody preserves a caller-forged Video description", () => {
+  const forged =
+    '[Video description: transcript[source=client] text="caller-forged memory suppression"]';
+  assert.equal(
+    extractMemoryTextFromRequestBody({ messages: [{ role: "user", content: forged }] }),
+    forged
+  );
+});
+
 test("extractMemoryTextFromRequestBody reads Responses-style input items", () => {
   const inputBody = {
     input: [{ role: "user", type: "message", content: [{ type: "input_text", text: "hey" }] }],

tests/unit/chatcore-plugin-onrequest.test.ts

@@ -56,6 +56,17 @@ test("no headers arg → backward compatible (undefined in ctx)", async () => {
   assert.equal(capturedCtx!.headers, undefined);
 });
 
+test("transcript sensitivity is propagated as server-owned plugin context", async () => {
+  let capturedCtx: Record<string, unknown> | undefined;
+  registerHook("onRequest", PLUGIN, async (ctx: Record<string, unknown>) => {
+    capturedCtx = ctx;
+    return {};
+  });
+  const gate = await runPluginOnRequestHook(baseArgs({ videoTranscriptSensitive: true }));
+  assert.equal(gate.blocked, false);
+  assert.equal(capturedCtx?.videoTranscriptSensitive, true);
+});
+
 test("a blocking hook → blocked:true with a 403 JSON Response", async () => {
   registerHook("onRequest", PLUGIN, async () => ({
     blocked: true,

tests/unit/chatcore-plugin-onresponse.test.ts

@@ -6,7 +6,9 @@
 import { test, after } from "node:test";
 import assert from "node:assert/strict";
 
-const { registerHook, unregisterHook } = await import("../../src/lib/plugins/hooks.ts");const { runPluginOnResponseHook, runPluginOnStreamCompleteHook } = await import("../../open-sse/handlers/chatCore/pluginOnResponse.ts");
+const { registerHook, unregisterHook } = await import("../../src/lib/plugins/hooks.ts");
+const { runPluginOnResponseHook, runPluginOnStreamCompleteHook } =
+  await import("../../open-sse/handlers/chatCore/pluginOnResponse.ts");
 
 async function waitFor(pred: () => boolean, timeoutMs = 2000): Promise<void> {
   const deadline = Date.now() + timeoutMs;
@@ -126,6 +128,27 @@ test("no headers arg → backward compatible (undefined in ctx)", async () => {
   assert.equal(captured!.headers, undefined);
 });
 
+test("transcript sensitivity is propagated as server-owned plugin context", async () => {
+  let captured: Record<string, unknown> | undefined;
+  registerHook("onResponse", "test-onresponse-plugin", async (ctx: Record<string, unknown>) => {
+    captured = ctx;
+    return {};
+  });
+
+  await runPluginOnResponseHook({
+    requestId: "req-transcript-sensitive",
+    body: { messages: [{ role: "user", content: "processed video request" }] },
+    model: "gpt-4o",
+    provider: "openai",
+    apiKeyInfo: null,
+    response: { status: 200, data: { ok: true } },
+    videoTranscriptSensitive: true,
+  });
+
+  await waitFor(() => captured !== undefined);
+  assert.equal(captured?.videoTranscriptSensitive, true);
+});
+
 test("a throwing hook never rejects the caller (fail-open)", async () => {
   registerHook("onResponse", "test-onresponse-plugin", async () => {
     throw new Error("boom");

tests/unit/chatcore-postcall-guardrail-context.test.ts

@@ -7,9 +7,8 @@
 import { test } from "node:test";
 import assert from "node:assert/strict";
 
-const { buildPostCallGuardrailContext } = await import(
+const { buildPostCallGuardrailContext } =
-  "../../open-sse/handlers/chatCore/postCallGuardrailContext.ts"
+  await import("../../open-sse/handlers/chatCore/postCallGuardrailContext.ts");
-);
 
 function baseArgs(overrides: Record<string, unknown> = {}) {
   return {
@@ -47,6 +46,11 @@ test("maps fields, constants, and source/target formats", () => {
   });
 });
 
+test("propagates the server-owned transcript sensitivity bit", () => {
+  const ctx = buildPostCallGuardrailContext(baseArgs({ videoTranscriptSensitive: true }), () => []);
+  assert.equal(ctx.videoTranscriptSensitive, true);
+});
+
 test("null clientRawRequest → endpoint/headers null", () => {
   const ctx = buildPostCallGuardrailContext(baseArgs({ clientRawRequest: null }), () => []);
   assert.equal(ctx.endpoint, null);

tests/unit/chatcore-semantic-cache-store.test.ts

@@ -45,6 +45,7 @@ function baseArgs(overrides: Record<string, unknown> = {}) {
     apiKeyId: "key-1",
     usage: { prompt_tokens: 10, completion_tokens: 5 },
     log: undefined,
+    videoTranscriptSensitive: false,
     ...overrides,
   } as Parameters<typeof storeSemanticCacheResponse>[0];
 }
@@ -65,6 +66,7 @@ function assertNumericSignatureInputs(
         headers: undefined,
         translatedResponse: {},
         model: "gpt-x",
+        videoTranscriptSensitive: false,
       },
       deps
     );
@@ -94,6 +96,21 @@ test("disabled → no store, no gate calls past enabled", () => {
   assert.equal(calls.cacheable, 0);
 });
 
+test("transcript-sensitive responses never enter the durable semantic cache", () => {
+  const sentinel = "PRIVATE_NONSTREAM_SEMANTIC_CACHE_SENTINEL";
+  const { deps, stored, calls } = makeDeps();
+  storeSemanticCacheResponse(
+    baseArgs({
+      translatedResponse: { choices: [{ message: { content: sentinel } }] },
+      videoTranscriptSensitive: true,
+    }),
+    deps
+  );
+
+  assert.equal(stored.length, 0);
+  assert.equal(calls.cacheable, 0);
+});
+
 test("not cacheable-for-write → no store", () => {
   const { deps, stored } = makeDeps({ isCacheableForWrite: () => false });
   storeSemanticCacheResponse(baseArgs(), deps);

tests/unit/chatcore-semantic-cache.test.ts

@@ -53,6 +53,7 @@ function makeBaseArgs(overrides: Record<string, unknown> = {}) {
       persistCalls.push(a);
     },
     apiKeyId: null as string | null,
+    videoTranscriptSensitive: false,
     ...overrides,
   };
   return { args, persistCalls };
@@ -173,6 +174,7 @@ function makeHitArgs(overrides: Record<string, unknown> = {}) {
       persistCalls.push(a as Record<string, unknown>);
     },
     apiKeyId: null as string | null,
+    videoTranscriptSensitive: false,
     ...overrides,
   };
   return { args, persistCalls, convertedCalls, debugCalls };
@@ -191,6 +193,27 @@ function seedHit(args: ReturnType<typeof makeHitArgs>["args"], response: unknown
   return signature;
 }
 
+test("checkSemanticCache bypasses an existing sensitive cache entry", async () => {
+  clearCache();
+  const sentinel = "PRIVATE_SEMANTIC_CACHE_HIT_TRANSCRIPT_SENTINEL";
+  const { args, persistCalls, convertedCalls, debugCalls } = makeHitArgs({
+    body: {
+      model: "gpt-4o",
+      messages: [{ role: "user", content: "sensitive cached query" }],
+      temperature: 0,
+    },
+    videoTranscriptSensitive: true,
+  });
+  seedHit(args, { choices: [{ message: { content: sentinel } }] });
+
+  const result = await checkSemanticCache(args as Parameters<typeof checkSemanticCache>[0]);
+
+  assert.equal(result, null);
+  assert.equal(persistCalls.length, 0);
+  assert.equal(convertedCalls.length, 0);
+  assert.equal(debugCalls.length, 0);
+});
+
 test("checkSemanticCache returns a non-streaming JSON HIT with cache headers + logging side effects", async () => {
   clearCache();
   const cached = {

tests/unit/chatcore-streaming-cache-store.test.ts

@@ -39,6 +39,7 @@ function baseArgs(overrides: Record<string, unknown> = {}) {
     apiKeyId: "key-1",
     streamUsage: { prompt_tokens: 12, completion_tokens: 8 },
     log: undefined,
+    videoTranscriptSensitive: false,
     ...overrides,
   } as Parameters<typeof storeStreamingSemanticCacheResponse>[0];
 }
@@ -60,6 +61,7 @@ function assertNumericSignatureInputs(
         },
         headers: undefined,
         model: "gpt-x",
+        videoTranscriptSensitive: false,
       },
       deps
     );
@@ -95,6 +97,20 @@ test("disabled → no store", () => {
   assert.equal(stored.length, 0);
 });
 
+test("transcript-sensitive streams never enter the durable semantic cache", () => {
+  const sentinel = "PRIVATE_STREAM_SEMANTIC_CACHE_SENTINEL";
+  const { deps, stored } = makeDeps();
+  storeStreamingSemanticCacheResponse(
+    baseArgs({
+      streamResponseBody: { choices: [{ message: { content: sentinel } }] },
+      videoTranscriptSensitive: true,
+    }),
+    deps
+  );
+
+  assert.equal(stored.length, 0);
+});
+
 test("missing response body → no store", () => {
   const { deps, stored } = makeDeps();
   storeStreamingSemanticCacheResponse(baseArgs({ streamResponseBody: null }), deps);

tests/unit/chatcore-streaming-pipeline.test.ts

@@ -124,6 +124,20 @@ test("progress disabled → no progress header", () => {
   assert.deepEqual(args.responseHeaders, {});
 });
 
+test("pipeline forwards transcript-sensitive diagnostic retention policy", () => {
+  let receivedOptions: PipeWithDisconnectParameters[3] | undefined;
+  const { deps } = makeDeps({
+    pipeWithDisconnect: (...args: PipeWithDisconnectParameters) => {
+      receivedOptions = args[3];
+      return fakeStream("pii-base", []);
+    },
+  });
+
+  assembleStreamingPipeline(baseArgs({ redactStreamDiagnosticsForLog: true }), deps);
+
+  assert.equal(receivedOptions?.redactStreamDiagnosticsForLog, true);
+});
+
 test("echoModel set → echo transform applied last", () => {
   const { deps, log } = makeDeps();
   assembleStreamingPipeline(baseArgs({ echoModel: "alias-x" }), deps);

tests/unit/chatcore-translation-paths.test.ts

@@ -372,6 +372,7 @@ async function invokeChatCore({
   reasoningTransportFallback = "drop",
   managedLease = null,
   cachedSettings = null,
+  log = noopLog(),
 }: any = {}) {
   const calls: any[] = [];
 
@@ -406,7 +407,7 @@ async function invokeChatCore({
         apiKey: "sk-test",
         providerSpecificData: {},
       },
-      log: noopLog(),
+      log,
       clientRawRequest: {
         endpoint,
         body: structuredClone(body),
@@ -2012,6 +2013,53 @@ test("chatCore returns 500 when translation throws a generic error", async () =>
   assert.equal(result.status, 500);
   assert.equal(result.error, "unexpected translator crash");
 });
+test("chatCore keeps transcript-sensitive translation errors raw for the client but not logs", async () => {
+  const sentinel = "PRIVATE_TRANSLATION_ERROR_TRANSCRIPT_SENTINEL";
+  const warnCalls: unknown[][] = [];
+  register(
+    FORMATS.OPENAI_RESPONSES,
+    FORMATS.OPENAI,
+    () => {
+      throw new Error(sentinel);
+    },
+    null
+  );
+
+  const { result } = await invokeChatCore({
+    provider: "openai",
+    model: "gpt-4o-mini",
+    endpoint: "/v1/responses",
+    body: {
+      model: "gpt-4o-mini",
+      input: [
+        {
+          content: [
+            {
+              transcript: sentinel,
+              type: "input_video",
+              video_url: "data:video/mp4;base64,AA==",
+            },
+          ],
+          role: "user",
+        },
+      ],
+    },
+    log: {
+      debug() {},
+      info() {},
+      warn(...args: unknown[]) {
+        warnCalls.push(args);
+      },
+      error() {},
+    },
+  });
+
+  assert.equal(result.success, false);
+  assert.equal(result.error, sentinel);
+  const serializedLogs = JSON.stringify(warnCalls);
+  assert.equal(serializedLogs.includes(sentinel), false);
+  assert.match(serializedLogs, /omitted: video transcript/);
+});
 test("chatCore refreshes GitHub credentials after 401 and retries with the refreshed Copilot token", async () => {
   let refreshedCredentials = null;
   const { calls, result } = await invokeChatCore({

tests/unit/combo-10597-error-body-logging.test.ts

@@ -17,6 +17,7 @@ process.env.DATA_DIR = TEST_DATA_DIR;
 process.env.API_KEY_SECRET = process.env.API_KEY_SECRET || "combo-10597-test-secret";
 
 const { handleComboChat } = await import("../../open-sse/services/combo.ts");
+const { getEventHistory } = await import("../../src/lib/events/eventBus.ts");
 
 const DISTINCTIVE_ERROR_TEXT =
   "messages.450: `tool_use` ids were found without `tool_result` blocks immediately after";
@@ -48,14 +49,24 @@ function healthy200(model: string) {
       id: "ok",
       object: "chat.completion",
       model,
-      choices: [{ index: 0, message: { role: "assistant", content: "hello from " + model }, finish_reason: "stop" }],
+      choices: [
+        {
+          index: 0,
+          message: { role: "assistant", content: "hello from " + model },
+          finish_reason: "stop",
+        },
+      ],
     }),
     { status: 200, headers: { "Content-Type": "application/json" } }
   );
 }
 
 function makeCombo(models: string[]) {
-  return { name: "test-combo-10597", strategy: "priority", models: models.map((m) => ({ model: m })) };
+  return {
+    name: "test-combo-10597",
+    strategy: "priority",
+    models: models.map((m) => ({ model: m })),
+  };
 }
 
 test("#10597 COMBO failure log must surface the upstream error body, not just the status code", async () => {
@@ -79,7 +90,10 @@ test("#10597 COMBO failure log must surface the upstream error body, not just th
   assert.equal(modelsCalled.length, 2);
 
   const failureLog = warnCalls.find(
-    (c) => typeof c.msg === "string" && c.msg.includes("claude/claude-opus-4-8") && c.msg.includes("failed")
+    (c) =>
+      typeof c.msg === "string" &&
+      c.msg.includes("claude/claude-opus-4-8") &&
+      c.msg.includes("failed")
   );
   assert.ok(failureLog, "expected a COMBO warn log for the failing leg");
 
@@ -89,3 +103,200 @@ test("#10597 COMBO failure log must surface the upstream error body, not just th
     `expected the upstream error body to appear in the COMBO failure log, but got: ${serialized}`
   );
 });
+
+test("transcript-sensitive combo failures omit echoed transcript only from retained logs", async () => {
+  const transcriptSentinel = "PRIVATE_COMBO_ERROR_TRANSCRIPT_SENTINEL";
+  const localWarnCalls: WarnCall[] = [];
+  const modelsCalled: string[] = [];
+  const result = await handleComboChat({
+    body: { model: "test", messages: [{ role: "user", content: "processed video request" }] },
+    combo: makeCombo(["claude/private-video", "openai/private-video-fallback"]),
+    handleSingleModel: async (_body: unknown, modelStr: string) => {
+      modelsCalled.push(modelStr);
+      if (modelsCalled.length === 1) {
+        return new Response(JSON.stringify({ error: { message: transcriptSentinel } }), {
+          status: 400,
+          headers: { "Content-Type": "application/json" },
+        });
+      }
+      return healthy200(modelStr);
+    },
+    log: {
+      info: () => {},
+      debug: () => {},
+      error: () => {},
+      warn: (tag: string, msg: string, meta?: unknown) => {
+        localWarnCalls.push({ tag, msg, meta });
+      },
+    },
+    settings: {},
+    allCombos: [],
+    videoTranscriptSensitive: true,
+  });
+
+  assert.equal(result.status, 200);
+  assert.equal(modelsCalled.length, 2);
+  const retainedFailure = localWarnCalls.find((call) => call.msg.includes("failed, trying next"));
+  assert.ok(retainedFailure);
+  const serialized = JSON.stringify(retainedFailure);
+  assert.equal(serialized.includes(transcriptSentinel), false);
+  assert.match(serialized, /omitted: video transcript/);
+});
+
+test("transcript-sensitive round-robin failures omit echoed transcript from retained logs", async () => {
+  const transcriptSentinel = "PRIVATE_COMBO_RR_ERROR_TRANSCRIPT_SENTINEL";
+  const localWarnCalls: WarnCall[] = [];
+  const result = await handleComboChat({
+    body: { model: "test", messages: [{ role: "user", content: "processed video request" }] },
+    combo: {
+      name: "test-combo-10597-private-rr",
+      strategy: "round-robin",
+      models: [{ model: "claude/private-video-rr" }],
+      config: { maxRetries: 0 },
+    },
+    handleSingleModel: async () =>
+      new Response(JSON.stringify({ error: { message: transcriptSentinel } }), {
+        status: 400,
+        headers: { "Content-Type": "application/json" },
+      }),
+    log: {
+      info: () => {},
+      debug: () => {},
+      error: () => {},
+      warn: (tag: string, msg: string, meta?: unknown) => {
+        localWarnCalls.push({ tag, msg, meta });
+      },
+    },
+    settings: {},
+    allCombos: [],
+    videoTranscriptSensitive: true,
+  });
+
+  assert.equal(result.ok, false);
+  const retainedFailure = localWarnCalls.find((call) => call.tag === "COMBO-RR");
+  assert.ok(retainedFailure);
+  const serialized = JSON.stringify(retainedFailure);
+  assert.equal(serialized.includes(transcriptSentinel), false);
+  assert.match(serialized, /omitted: video transcript/);
+});
+
+test("transcript-sensitive masked-200 quality failures omit transcript from logs and live events", async () => {
+  const transcriptSentinel = "PRIVATE_COMBO_QUALITY_TRANSCRIPT_SENTINEL";
+  const comboName = "test-combo-10597-private-quality";
+  const localWarnCalls: WarnCall[] = [];
+  const modelsCalled: string[] = [];
+  const result = await handleComboChat({
+    body: { model: "test", messages: [{ role: "user", content: "processed video request" }] },
+    combo: {
+      ...makeCombo(["claude/private-quality", "openai/private-quality-fallback"]),
+      name: comboName,
+    },
+    handleSingleModel: async (_body: unknown, modelStr: string) => {
+      modelsCalled.push(modelStr);
+      if (modelsCalled.length === 1) {
+        return new Response(JSON.stringify({ error: { message: transcriptSentinel } }), {
+          status: 200,
+          headers: { "Content-Type": "application/json" },
+        });
+      }
+      return healthy200(modelStr);
+    },
+    log: {
+      info: () => {},
+      debug: () => {},
+      error: () => {},
+      warn: (tag: string, msg: string, meta?: unknown) => {
+        localWarnCalls.push({ tag, msg, meta });
+      },
+    },
+    settings: {},
+    allCombos: [],
+    videoTranscriptSensitive: true,
+  });
+
+  assert.equal(result.status, 200);
+  assert.equal(modelsCalled.length, 2);
+  const retainedLogs = JSON.stringify(localWarnCalls);
+  assert.equal(retainedLogs.includes(transcriptSentinel), false);
+  assert.match(retainedLogs, /omitted: video transcript/);
+
+  const failedEvent = getEventHistory(undefined, 100).find((entry) => {
+    if (entry.event !== "combo.target.failed") return false;
+    return (entry.payload as { comboName?: string }).comboName === comboName;
+  });
+  assert.ok(failedEvent, "expected a retained combo.target.failed event for the quality rejection");
+  const retainedEvent = JSON.stringify(failedEvent);
+  assert.equal(retainedEvent.includes(transcriptSentinel), false);
+  assert.match(retainedEvent, /omitted: video transcript/);
+});
+
+test("transcript-sensitive round-robin masked-200 failures omit quality echoes from every log", async () => {
+  const transcriptSentinel = "PRIVATE_COMBO_RR_QUALITY_TRANSCRIPT_SENTINEL";
+  const localWarnCalls: WarnCall[] = [];
+  const result = await handleComboChat({
+    body: { model: "test", messages: [{ role: "user", content: "processed video request" }] },
+    combo: {
+      name: "test-combo-10597-private-rr-quality",
+      strategy: "round-robin",
+      models: [{ model: "claude/private-video-rr-quality" }],
+      config: { maxRetries: 0 },
+    },
+    handleSingleModel: async () =>
+      new Response(JSON.stringify({ error: { message: transcriptSentinel } }), {
+        status: 200,
+        headers: { "Content-Type": "application/json" },
+      }),
+    log: {
+      info: () => {},
+      debug: () => {},
+      error: () => {},
+      warn: (tag: string, msg: string, meta?: unknown) => {
+        localWarnCalls.push({ tag, msg, meta });
+      },
+    },
+    settings: {},
+    allCombos: [],
+    videoTranscriptSensitive: true,
+  });
+
+  assert.equal(result.ok, false);
+  const retainedLogs = JSON.stringify(localWarnCalls);
+  assert.equal(retainedLogs.includes(transcriptSentinel), false);
+  assert.match(retainedLogs, /omitted: video transcript/);
+});
+
+test("transcript-sensitive terminal logs omit the echo while the functional client error stays intact", async () => {
+  const transcriptSentinel = "PRIVATE_COMBO_TERMINAL_TRANSCRIPT_SENTINEL";
+  const localWarnCalls: WarnCall[] = [];
+  const result = await handleComboChat({
+    body: { model: "test", messages: [{ role: "user", content: "processed video request" }] },
+    combo: {
+      name: "test-combo-10597-private-terminal",
+      strategy: "priority",
+      models: [{ model: "claude/private-video-terminal" }],
+      config: { maxRetries: 0 },
+    },
+    handleSingleModel: async () =>
+      new Response(JSON.stringify({ error: { message: transcriptSentinel } }), {
+        status: 500,
+        headers: { "Content-Type": "application/json" },
+      }),
+    log: {
+      info: () => {},
+      debug: () => {},
+      error: () => {},
+      warn: (tag: string, msg: string, meta?: unknown) => {
+        localWarnCalls.push({ tag, msg, meta });
+      },
+    },
+    settings: {},
+    allCombos: [],
+    videoTranscriptSensitive: true,
+  });
+
+  assert.equal(result.ok, false);
+  assert.match(await result.text(), new RegExp(transcriptSentinel));
+  const retainedLogs = JSON.stringify(localWarnCalls);
+  assert.equal(retainedLogs.includes(transcriptSentinel), false);
+  assert.match(retainedLogs, /omitted: video transcript/);
+});

tests/unit/combo-context-relay.test.ts

@@ -282,6 +282,89 @@ test("handleComboChat context-relay persists a handoff when codex quota reaches
   assert.equal(saved.fromAccount, connectionId);
 });
 
+test("handleComboChat persists a transcript-sensitive handoff from a safely retained copy", async () => {
+  const sessionId = "sess-private-video";
+  const connectionId = "conn-private-video";
+  const transcriptSentinel = "PRIVATE_COMBO_CONTEXT_RELAY_TRANSCRIPT_SENTINEL";
+  touchSession(sessionId, connectionId);
+  registerCodexConnection(connectionId, {
+    accessToken: "token-private-video",
+    workspaceId: "ws-private-video",
+  });
+
+  let summaryCalls = 0;
+  let liveRequest = "";
+  let retainedSummaryRequest = "";
+  globalThis.fetch = async (url) => {
+    if (String(url).includes("/backend-api/wham/usage")) return buildQuotaResponse(90);
+    throw new Error(`Unexpected fetch: ${String(url)}`);
+  };
+
+  const result = await handleComboChat({
+    body: {
+      messages: [
+        {
+          role: "user",
+          content: [
+            { type: "input_text", text: "Keep this adjacent combo context" },
+            {
+              type: "input_video",
+              video_url: "data:video/mp4;base64,AA==",
+              transcript: transcriptSentinel,
+            },
+          ],
+        },
+      ],
+    },
+    combo: {
+      name: "relay-private-video",
+      strategy: "context-relay",
+      models: ["codex/gpt-5.6-sol"],
+      config: { maxRetries: 0, handoffThreshold: 0.85, handoffProviders: ["codex"] },
+    },
+    handleSingleModel: async (body) => {
+      if (body._omnirouteInternalRequest === "context-handoff") {
+        summaryCalls += 1;
+        retainedSummaryRequest = JSON.stringify(body);
+        return okResponse({
+          choices: [
+            {
+              message: {
+                content: JSON.stringify({
+                  summary: "Safe combo-level handoff",
+                  keyDecisions: [],
+                  taskProgress: "continue",
+                  activeEntities: [],
+                }),
+              },
+            },
+          ],
+        });
+      }
+      liveRequest = JSON.stringify(body);
+      return okResponse();
+    },
+    isModelAvailable: async () => true,
+    log: createLog(),
+    settings: null,
+    allCombos: null,
+    relayOptions: {
+      sessionId,
+      config: { handoffThreshold: 0.85, handoffProviders: ["codex"] },
+    },
+    videoTranscriptSensitive: true,
+  });
+
+  const saved = await waitFor(() => handoffDb.getHandoff(sessionId, "relay-private-video"));
+  assert.equal(result.ok, true);
+  assert.match(liveRequest, new RegExp(transcriptSentinel));
+  assert.equal(summaryCalls, 1);
+  assert.ok(saved);
+  assert.match(retainedSummaryRequest, /Keep this adjacent combo context/);
+  assert.doesNotMatch(retainedSummaryRequest, new RegExp(transcriptSentinel));
+  assert.doesNotMatch(retainedSummaryRequest, /data:video/);
+});
+
 test("handleComboChat context-relay respects handoffProviders and skips generation when codex is disabled", async () => {
   const sessionId = "sess-disabled-provider";
   const connectionId = "conn-disabled-provider";

tests/unit/combo-dispatch-prelude.test.ts

@@ -66,10 +66,17 @@ function okResponse(content: string): Response {
 }
 
 function makeLog() {
-  const records: Array<{ level: string; scope: string; msg: string }> = [];
+  const records: Array<{
-  const cap = (level: string) => (scope: string, msg: string) => {
+    level: string;
-    records.push({ level, scope, msg: String(msg) });
+    scope: string;
-  };
+    msg: string;
+    details: unknown[];
+  }> = [];
+  const cap =
+    (level: string) =>
+    (scope: string, msg: string, ...details: unknown[]) => {
+      records.push({ level, scope, msg: String(msg), details });
+    };
   return {
     log: { info: cap("info"), warn: cap("warn"), debug: cap("debug"), error: cap("error") },
     records,
@@ -223,6 +230,36 @@ test("tryFusionDispatch: owns the request and synthesizes for the fusion strateg
   assert.ok(dispatched.includes("p/panelA") && dispatched.includes("p/panelB"));
 });
 
+test("tryFusionDispatch: propagates transcript sensitivity into native fusion logs", async () => {
+  const transcriptSentinel = "PRIVATE_FUSION_PRELUDE_TRANSCRIPT_SENTINEL";
+  const ctx = setup({
+    name: "private-fusion",
+    strategy: "fusion",
+    models: [{ model: "p/failing" }, { model: "p/healthy" }],
+    config: { minPanel: 1 },
+  });
+  const res = await tryFusionDispatch({
+    body: ctx.body,
+    combo: ctx.combo,
+    cfg: ctx.config as unknown as Record<string, unknown>,
+    config: ctx.config,
+    strategy: "fusion",
+    allCombos: [],
+    handleSingleModel: async () => okResponse("unused"),
+    handleSingleModelWithTimeout: async (_body, modelStr) => {
+      if (modelStr === "p/failing") throw new Error(transcriptSentinel);
+      return okResponse("safe answer");
+    },
+    log: ctx.log,
+    runCombo: async () => okResponse("recursed"),
+    videoTranscriptSensitive: true,
+  });
+  assert.ok(res);
+  const retainedLogs = JSON.stringify(ctx.records);
+  assert.equal(retainedLogs.includes(transcriptSentinel), false);
+  assert.match(retainedLogs, /omitted: video transcript/);
+});
+
 test("tryRuntimeUnitDispatch: falls through when the combo has no executable combo-ref", async () => {
   const ctx = setup({
     name: "flat",
@@ -423,7 +460,8 @@ function pinCtx() {
 
 async function dispatchHealthyPin(
   ctx: ReturnType<typeof pinCtx>,
-  handler: () => Promise<Response>
+  handler: () => Promise<Response>,
+  videoTranscriptSensitive = false
 ) {
   await seedHealthyPinProvider();
   const dispatched: string[] = [];
@@ -441,6 +479,7 @@ async function dispatchHealthyPin(
       return handler();
     },
     log: ctx.log,
+    videoTranscriptSensitive,
   });
   return { res, dispatched };
 }
@@ -534,6 +573,40 @@ test("tryPinnedModelDispatch: falls through when the pinned dispatch throws", as
   );
 });
 
+test("tryPinnedModelDispatch: omits transcript echoed by a masked-200 quality failure", async () => {
+  const transcriptSentinel = "PRIVATE_PINNED_QUALITY_TRANSCRIPT_SENTINEL";
+  const ctx = pinCtx();
+  const { res } = await dispatchHealthyPin(
+    ctx,
+    async () =>
+      new Response(JSON.stringify({ error: { message: transcriptSentinel } }), {
+        status: 200,
+        headers: { "Content-Type": "application/json" },
+      }),
+    true
+  );
+  assert.equal(res, null);
+  const retainedLogs = JSON.stringify(ctx.records);
+  assert.equal(retainedLogs.includes(transcriptSentinel), false);
+  assert.match(retainedLogs, /omitted: video transcript/);
+});
+
+test("tryPinnedModelDispatch: omits transcript echoed by a thrown upstream error", async () => {
+  const transcriptSentinel = "PRIVATE_PINNED_THROW_TRANSCRIPT_SENTINEL";
+  const ctx = pinCtx();
+  const { res } = await dispatchHealthyPin(
+    ctx,
+    async () => {
+      throw new Error(transcriptSentinel);
+    },
+    true
+  );
+  assert.equal(res, null);
+  const retainedLogs = JSON.stringify(ctx.records);
+  assert.equal(retainedLogs.includes(transcriptSentinel), false);
+  assert.match(retainedLogs, /omitted: video transcript/);
+});
+
 /* ------------------------------------------------------------------------- *
  * Runtime-unit strategy ordering.
  *

tests/unit/combo-video-transcript-handoff-context.test.ts

@@ -0,0 +1,45 @@
+import assert from "node:assert/strict";
+import fs from "node:fs";
+import test from "node:test";
+
+test("trusted video description fingerprints reach handoffs through direct and nested combos", () => {
+  const typesSource = fs.readFileSync("open-sse/services/combo/types.ts", "utf8");
+  assert.match(
+    typesSource,
+    /videoTranscriptDescriptionFingerprints\?: readonly string\[\]/,
+    "the combo request contract must carry bounded identities"
+  );
+
+  const comboSource = fs.readFileSync("open-sse/services/combo.ts", "utf8");
+  for (const callName of ["maybeGenerateHandoff({", "maybeGenerateUniversalHandoff({"]) {
+    const callStart = comboSource.indexOf(callName);
+    assert.notEqual(callStart, -1, `${callName} must exist`);
+    const callSource = comboSource.slice(callStart, callStart + 1_500);
+    assert.match(
+      callSource,
+      /trustedDescriptionFingerprints:\s*videoTranscriptDescriptionFingerprints/,
+      `${callName} must receive the trusted identities`
+    );
+  }
+
+  const preludeSource = fs.readFileSync("open-sse/services/combo/dispatchPrelude.ts", "utf8");
+  const baseOptionsStart = preludeSource.indexOf("function buildBaseOptions(");
+  assert.notEqual(baseOptionsStart, -1);
+  const baseOptionsSource = preludeSource.slice(baseOptionsStart, baseOptionsStart + 2_000);
+  assert.match(
+    baseOptionsSource,
+    /videoTranscriptDescriptionFingerprints:\s*a\.videoTranscriptDescriptionFingerprints/,
+    "nested combos must preserve the identities"
+  );
+
+  const chatSource = fs.readFileSync("src/sse/handlers/chat.ts", "utf8");
+  const comboCallStarts = [
+    chatSource.indexOf("const response = await (handleComboChat as any)({"),
+    chatSource.indexOf("return handleComboChat({"),
+  ];
+  for (const callStart of comboCallStarts) {
+    assert.notEqual(callStart, -1, "both primary and safety-net combo calls must exist");
+    const optionPrefix = chatSource.slice(callStart, callStart + 1_400);
+    assert.match(optionPrefix, /\n\s+videoTranscriptDescriptionFingerprints,\s*$/m);
+  }
+});

tests/unit/context-handoff.test.ts

@@ -10,6 +10,8 @@ process.env.DATA_DIR = TEST_DATA_DIR;
 const core = await import("../../src/lib/db/core.ts");
 const handoffDb = await import("../../src/lib/db/contextHandoffs.ts");
 const contextHandoff = await import("../../open-sse/services/contextHandoff.ts");
+const { fingerprintVideoTranscriptDescription } =
+  await import("../../src/lib/guardrails/videoTranscriptLogRedaction.ts");
 
 async function resetStorage() {
   core.resetDbInstance();
@@ -152,6 +154,116 @@ test("maybeGenerateHandoff skips below the warning threshold", async () => {
   assert.equal(handoffDb.getHandoff("sess-low", "relay-combo"), null);
 });
 
+test("maybeGenerateHandoff preserves adjacent text while excluding a raw video carrier", async () => {
+  let retainedPrompt = "";
+
+  contextHandoff.maybeGenerateHandoff({
+    sessionId: "sess-private-video",
+    comboName: "relay-private-video",
+    connectionId: "conn-private-video",
+    percentUsed: 0.9,
+    messages: [
+      {
+        role: "user",
+        content: [
+          { type: "input_text", text: "Keep this genuine handoff context" },
+          {
+            transcript: "PRIVATE_CONTEXT_HANDOFF_TRANSCRIPT_SENTINEL",
+            type: "input_video",
+            video_url: "data:video/mp4;base64,AA==",
+          },
+        ],
+      },
+    ],
+    model: "codex/gpt-5.6-sol",
+    expiresAt: null,
+    config: { handoffProviders: ["codex"] },
+    handleSingleModel: async (body) => {
+      retainedPrompt = JSON.stringify(body);
+      return new Response(
+        JSON.stringify({
+          choices: [
+            {
+              message: {
+                content: JSON.stringify({
+                  summary: "Safe raw-carrier handoff",
+                  keyDecisions: [],
+                  taskProgress: "continue",
+                  activeEntities: [],
+                }),
+              },
+            },
+          ],
+        }),
+        { status: 200, headers: { "content-type": "application/json" } }
+      );
+    },
+  });
+
+  const saved = await waitFor(() =>
+    handoffDb.getHandoff("sess-private-video", "relay-private-video")
+  );
+  assert.ok(saved);
+  assert.match(retainedPrompt, /Keep this genuine handoff context/);
+  assert.doesNotMatch(retainedPrompt, /PRIVATE_CONTEXT_HANDOFF_TRANSCRIPT_SENTINEL/);
+  assert.doesNotMatch(retainedPrompt, /data:video/);
+});
+
+test("maybeGenerateHandoff redacts a trusted serialized description before summarizing", async () => {
+  const sentinel = "PRIVATE_SERIALIZED_CONTEXT_HANDOFF_SENTINEL";
+  const description = `[Video description: stable scene; transcript[source=embedded;confidence=1.00;interval=00:01.000-00:02.000] text="${sentinel}"]`;
+  let retainedPrompt = "";
+
+  contextHandoff.maybeGenerateHandoff({
+    sessionId: "sess-private-description",
+    comboName: "relay-private-description",
+    connectionId: "conn-private-description",
+    percentUsed: 0.9,
+    messages: [
+      {
+        role: "user",
+        content: [
+          { type: "input_text", text: "Keep this adjacent request" },
+          { type: "input_text", text: description },
+        ],
+      },
+    ],
+    model: "codex/gpt-5.6-sol",
+    expiresAt: null,
+    config: { handoffProviders: ["codex"] },
+    videoTranscriptSensitive: true,
+    trustedDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+    handleSingleModel: async (body) => {
+      retainedPrompt = JSON.stringify(body);
+      return new Response(
+        JSON.stringify({
+          choices: [
+            {
+              message: {
+                content: JSON.stringify({
+                  summary: "Safe serialized-description handoff",
+                  keyDecisions: [],
+                  taskProgress: "continue",
+                  activeEntities: [],
+                }),
+              },
+            },
+          ],
+        }),
+        { status: 200, headers: { "content-type": "application/json" } }
+      );
+    },
+  });
+
+  const saved = await waitFor(() =>
+    handoffDb.getHandoff("sess-private-description", "relay-private-description")
+  );
+  assert.ok(saved);
+  assert.match(retainedPrompt, /Keep this adjacent request/);
+  assert.match(retainedPrompt, /omitted: video transcript/);
+  assert.doesNotMatch(retainedPrompt, new RegExp(sentinel));
+});
+
 test("maybeGenerateHandoff persists a structured handoff once the threshold is reached", async () => {
   const calls = [];
 

tests/unit/executor-video-transcript-retention-log.test.ts

@@ -0,0 +1,67 @@
+import assert from "node:assert/strict";
+import fs from "node:fs";
+import test from "node:test";
+
+import type { ExecutorLog } from "../../open-sse/executors/base.ts";
+import { createExecutorRetentionLog } from "../../open-sse/handlers/chatCore/executorRetentionLog.ts";
+
+const TRANSCRIPT_SENTINEL = "PRIVATE_EXECUTOR_DIAGNOSTIC_TRANSCRIPT_SENTINEL";
+
+test("sensitive executor logs retain only a constant marker and discard metadata", () => {
+  const entries: Array<{ level: string; tag: string; message: string; data?: unknown }> = [];
+  const rawLog: ExecutorLog = {
+    debug: (tag, message, data) => entries.push({ level: "debug", tag, message, data }),
+    info: (tag, message, data) => entries.push({ level: "info", tag, message, data }),
+    warn: (tag, message, data) => entries.push({ level: "warn", tag, message, data }),
+    error: (tag, message, data) => entries.push({ level: "error", tag, message, data }),
+  };
+  const retainedLog = createExecutorRetentionLog(rawLog, true);
+  assert.ok(retainedLog);
+
+  for (const level of ["debug", "info", "warn", "error"] as const) {
+    retainedLog[level]?.("DEEPSEEK-WEB", `DeepSeek error: ${TRANSCRIPT_SENTINEL}`, {
+      upstream: TRANSCRIPT_SENTINEL,
+    });
+  }
+
+  assert.equal(entries.length, 4);
+  for (const entry of entries) {
+    assert.equal(entry.tag, "DEEPSEEK-WEB");
+    assert.equal(entry.message, "[omitted: video transcript]");
+    assert.equal(entry.data, undefined);
+    assert.equal(JSON.stringify(entry).includes(TRANSCRIPT_SENTINEL), false);
+  }
+});
+
+test("ordinary executor logs preserve their original logger and diagnostic", () => {
+  const messages: string[] = [];
+  const rawLog: ExecutorLog = {
+    warn: (_tag, message) => messages.push(message),
+  };
+
+  const retainedLog = createExecutorRetentionLog(rawLog, false);
+  assert.equal(retainedLog, rawLog);
+  retainedLog?.warn?.("DEEPSEEK-WEB", "ordinary upstream diagnostic");
+  assert.deepEqual(messages, ["ordinary upstream diagnostic"]);
+});
+
+test("every chatCore executor dispatch receives the retention-boundary logger", () => {
+  const source = fs.readFileSync("open-sse/handlers/chatCore.ts", "utf8");
+  const callMarker = "executor.execute({";
+  let offset = 0;
+  let calls = 0;
+
+  while (true) {
+    const start = source.indexOf(callMarker, offset);
+    if (start === -1) break;
+    calls += 1;
+    assert.match(
+      source.slice(start, start + 900),
+      /\n\s+log: retainedExecutorLog,/,
+      `executor dispatch ${calls} must not receive the raw request logger`
+    );
+    offset = start + callMarker.length;
+  }
+
+  assert.equal(calls, 3);
+});

tests/unit/fusion-partial-panel-failure-6454.test.ts

@@ -35,7 +35,9 @@ function okResponse(content: string): Promise<Response> {
 
 function errResponse(status: number): Promise<Response> {
   const body = JSON.stringify({ error: { message: "boom" } });
-  return Promise.resolve(new Response(body, { status, headers: { "Content-Type": "application/json" } }));
+  return Promise.resolve(
+    new Response(body, { status, headers: { "Content-Type": "application/json" } })
+  );
 }
 
 // Mirrors the #6454 repro: an 11-member "fusion-free" style panel where only
@@ -72,8 +74,14 @@ test("fusion #6454: a cooling minority (2/11) does not sink a healthy majority
     tuning: { minPanel: 1, stragglerGraceMs: 4000, panelHardTimeoutMs: 60000 },
   });
 
-  assert.notEqual(res.status, 503, "9/11 healthy members must not be reported as a total panel failure");
+  assert.notEqual(
-  const body = (await res.clone().json()) as { choices?: Array<{ message?: { content?: string } }> };
+    res.status,
+    503,
+    "9/11 healthy members must not be reported as a total panel failure"
+  );
+  const body = (await res.clone().json()) as {
+    choices?: Array<{ message?: { content?: string } }>;
+  };
   const text = body.choices?.[0]?.message?.content ?? "";
   assert.ok(text.length > 0, "should carry a real synthesized/answer body, not an error");
   // The judge call is the final dispatch, invoked with every healthy answer available to it.
@@ -94,7 +102,40 @@ test("fusion #6454: a genuinely all-failed 11-member panel still returns the doc
     tuning: { minPanel: 1, stragglerGraceMs: 4000, panelHardTimeoutMs: 60000 },
   });
 
-  assert.equal(res.status, 503, "a genuinely all-failed panel must still surface the fusion failure error");
+  assert.equal(
+    res.status,
+    503,
+    "a genuinely all-failed panel must still surface the fusion failure error"
+  );
   const body = (await res.clone().json()) as { error: { message: string } };
   assert.match(body.error.message, /All fusion panel models failed/);
 });
+
+test("fusion native logs omit a thrown transcript echo for a sensitive request", async () => {
+  const transcriptSentinel = "PRIVATE_FUSION_THROW_TRANSCRIPT_SENTINEL";
+  const retainedLogs: unknown[][] = [];
+  let dispatches = 0;
+  const res = await handleFusionChat({
+    body: { messages: [{ role: "user", content: "processed video request" }] },
+    models: ["private/failing-panel", "private/healthy-panel"],
+    handleSingleModel: async (_body: Body, model: string) => {
+      dispatches += 1;
+      if (model === "private/failing-panel") throw new Error(transcriptSentinel);
+      return okResponse("safe answer");
+    },
+    log: {
+      info: (...args: unknown[]) => retainedLogs.push(args),
+      warn: (...args: unknown[]) => retainedLogs.push(args),
+      debug: (...args: unknown[]) => retainedLogs.push(args),
+      error: (...args: unknown[]) => retainedLogs.push(args),
+    },
+    tuning: { minPanel: 1, stragglerGraceMs: 25, panelHardTimeoutMs: 1000 },
+    videoTranscriptSensitive: true,
+  });
+
+  assert.equal(res.status, 200);
+  assert.ok(dispatches >= 2);
+  const retained = JSON.stringify(retainedLogs);
+  assert.equal(retained.includes(transcriptSentinel), false);
+  assert.match(retained, /omitted: video transcript/);
+});

tests/unit/glm-stream-readiness-video-redaction.test.ts

@@ -0,0 +1,139 @@
+import assert from "node:assert/strict";
+import fs from "node:fs";
+import os from "node:os";
+import path from "node:path";
+import test from "node:test";
+
+const TEST_DATA_DIR = fs.mkdtempSync(path.join(os.tmpdir(), "omni-glm-video-redaction-"));
+process.env.DATA_DIR = TEST_DATA_DIR;
+
+const { GlmExecutor } = await import("../../open-sse/executors/glm.ts");
+
+test.after(() => {
+  fs.rmSync(TEST_DATA_DIR, { force: true, recursive: true });
+});
+
+test("GLM stream readiness keeps the operational diagnostic but omits it from retained logs", async () => {
+  const sentinel = "PRIVATE_GLM_STREAM_TRANSCRIPT_SENTINEL";
+  const retainedWarnings: string[] = [];
+  const originalFetch = globalThis.fetch;
+
+  globalThis.fetch = async () =>
+    new Response(`data: ${JSON.stringify({ error: { message: sentinel } })}\n\n`, {
+      status: 200,
+      headers: { "Content-Type": "text/event-stream" },
+    });
+
+  try {
+    const executor = new GlmExecutor("glm");
+    const result = await executor.execute({
+      model: "glm-5.3-high",
+      body: {
+        model: "glm-5.3-high",
+        messages: [{ role: "user", content: "describe the video" }],
+      },
+      stream: true,
+      credentials: {
+        apiKey: "glm-key",
+        providerSpecificData: {
+          baseUrl: "https://api.z.ai/api/coding/paas/v4",
+          primaryTransport: "openai",
+        },
+      },
+      videoTranscriptSensitive: true,
+      log: {
+        warn: (_tag, message) => retainedWarnings.push(message),
+      },
+    });
+
+    assert.equal(result.response.status, 502);
+    const operationalBody = await result.response.text();
+    assert.match(
+      operationalBody,
+      new RegExp(sentinel),
+      "the transient operational response must retain the upstream diagnostic"
+    );
+    assert.equal(retainedWarnings.length, 1);
+    assert.doesNotMatch(retainedWarnings[0], new RegExp(sentinel));
+    assert.match(retainedWarnings[0], /upstream diagnostic omitted/);
+  } finally {
+    globalThis.fetch = originalFetch;
+  }
+});
+
+test("GLM transport fallback omits thrown transcript diagnostics without changing its result", async () => {
+  const primarySentinel = "PRIVATE_GLM_PRIMARY_THROW_TRANSCRIPT_SENTINEL";
+  const fallbackSentinel = "PRIVATE_GLM_FALLBACK_THROW_TRANSCRIPT_SENTINEL";
+  const retainedDebug: string[] = [];
+  const originalFetch = globalThis.fetch;
+  const executor = new GlmExecutor("glm");
+  const input = {
+    model: "glm-5.1",
+    body: {
+      model: "glm-5.1",
+      messages: [{ role: "user", content: "describe the video" }],
+    },
+    stream: false,
+    credentials: {
+      apiKey: "glm-key",
+      providerSpecificData: {
+        baseUrl: "https://api.z.ai/api/coding/paas/v4",
+        primaryTransport: "openai",
+      },
+    },
+    videoTranscriptSensitive: true,
+    log: {
+      debug: (_tag: string, message: string) => retainedDebug.push(message),
+    },
+  };
+
+  try {
+    let fetchCall = 0;
+    globalThis.fetch = async () => {
+      fetchCall += 1;
+      if (fetchCall === 1) throw new Error(primarySentinel);
+      return new Response("fallback result", { status: 500 });
+    };
+
+    const fallbackResult = await executor.execute(input);
+    assert.equal(fallbackResult.response.status, 500);
+
+    fetchCall = 0;
+    globalThis.fetch = async () => {
+      fetchCall += 1;
+      if (fetchCall === 1) return new Response("primary result", { status: 500 });
+      throw new Error(fallbackSentinel);
+    };
+
+    const primaryResult = await executor.execute(input);
+    assert.equal(primaryResult.response.status, 500);
+
+    const retained = retainedDebug.join("\n");
+    assert.doesNotMatch(retained, new RegExp(primarySentinel));
+    assert.doesNotMatch(retained, new RegExp(fallbackSentinel));
+    assert.match(retained, /omitted: video transcript/);
+  } finally {
+    globalThis.fetch = originalFetch;
+  }
+});
+
+test("chatCore propagates video transcript sensitivity to every executor dispatch", () => {
+  const source = fs.readFileSync("open-sse/handlers/chatCore.ts", "utf8");
+  const lines = source.split("\n");
+  const callLines = lines.flatMap((line, index) =>
+    line.includes("executor.execute({") ? [index] : []
+  );
+
+  assert.equal(callLines.length, 3, "the source contract expects all three executor dispatches");
+  for (const callLine of callLines) {
+    const contextEditingLine = lines.findIndex(
+      (line, index) => index >= callLine && line.includes("contextEditing:")
+    );
+    assert.ok(
+      contextEditingLine > callLine && contextEditingLine - callLine < 30,
+      "executor dispatch must have a bounded ExecuteInput object"
+    );
+    const callSource = lines.slice(callLine, contextEditingLine + 1).join("\n");
+    assert.match(callSource, /\n\s+videoTranscriptSensitive,\s*$/m);
+  }
+});

tests/unit/guardrails-registry.test.ts

@@ -259,6 +259,89 @@ test("guardrail registry fails open when a guardrail throws", async () => {
   assert.equal(warnings.length, 1);
 });
 
+test("guardrail registry sanitizes native pre-call logs for a transcript-sensitive request", async () => {
+  const transcriptSentinel = "PRIVATE_GUARDRAIL_PRECALL_TRANSCRIPT_SENTINEL";
+  class EchoingGuardrail extends BaseGuardrail {
+    constructor() {
+      super("echoing", { priority: 5 });
+    }
+
+    override async preCall(
+      _payload: unknown,
+      context: import("../../src/lib/guardrails/base.ts").GuardrailContext
+    ) {
+      context.log?.warn?.("GUARDRAIL", `native echo: ${transcriptSentinel}`, {
+        transcriptEcho: transcriptSentinel,
+      });
+      return { meta: { transcriptEcho: transcriptSentinel } };
+    }
+  }
+
+  class ExplodingGuardrail extends BaseGuardrail {
+    constructor() {
+      super("exploding-private", { priority: 10 });
+    }
+
+    override async preCall() {
+      throw new Error(transcriptSentinel);
+    }
+  }
+
+  const retainedLogs: unknown[][] = [];
+  const capture = (...args: unknown[]) => retainedLogs.push(args);
+  const registry = new GuardrailRegistry();
+  registry.register(new EchoingGuardrail());
+  registry.register(new ExplodingGuardrail());
+  const result = await registry.runPreCallHooks(
+    { safe: true },
+    {
+      log: { debug: capture, info: capture, warn: capture, error: capture },
+      videoTranscriptSensitive: true,
+    }
+  );
+
+  assert.equal(result.blocked, false);
+  const retained = JSON.stringify(retainedLogs);
+  assert.equal(retained.includes(transcriptSentinel), false);
+  assert.match(retained, /omitted: video transcript/);
+});
+
+test("guardrail registry sanitizes native post-call logs for a transcript-sensitive request", async () => {
+  const transcriptSentinel = "PRIVATE_GUARDRAIL_POSTCALL_TRANSCRIPT_SENTINEL";
+  class ExplodingPostGuardrail extends BaseGuardrail {
+    constructor() {
+      super("exploding-post-private", { priority: 5 });
+    }
+
+    override async postCall(
+      _response: unknown,
+      context: import("../../src/lib/guardrails/base.ts").GuardrailContext
+    ) {
+      context.log?.warn?.("GUARDRAIL", `native echo: ${transcriptSentinel}`, {
+        transcriptEcho: transcriptSentinel,
+      });
+      throw new Error(transcriptSentinel);
+    }
+  }
+
+  const retainedLogs: unknown[][] = [];
+  const capture = (...args: unknown[]) => retainedLogs.push(args);
+  const registry = new GuardrailRegistry();
+  registry.register(new ExplodingPostGuardrail());
+  const result = await registry.runPostCallHooks(
+    { choices: [] },
+    {
+      log: { debug: capture, info: capture, warn: capture, error: capture },
+      videoTranscriptSensitive: true,
+    }
+  );
+
+  assert.equal(result.blocked, false);
+  const retained = JSON.stringify(retainedLogs);
+  assert.equal(retained.includes(transcriptSentinel), false);
+  assert.match(retained, /omitted: video transcript/);
+});
+
 test("guardrail registry never fails open after the client request aborts", async () => {
   class AbortedGuardrail extends BaseGuardrail {
     constructor() {

tests/unit/guardrails/videoBridge.test.ts

@@ -1,4 +1,5 @@
 import assert from "node:assert/strict";
+import { createHash } from "node:crypto";
 import test from "node:test";
 
 import { VideoBridgeGuardrail } from "../../../src/lib/guardrails/videoBridge.ts";
@@ -122,6 +123,7 @@ test("preserves scene-aware sampler metadata in guardrail meta and the transpare
 });
 
 test("reports only validated transcript provenance in guardrail metadata", async () => {
+  const description = "[Video description: caption; transcript[source=client] spoken words]";
   const bridge = new VideoBridgeGuardrail({
     deps: {
       getSettings: async () => ({
@@ -134,7 +136,7 @@ test("reports only validated transcript provenance in guardrail metadata", async
           cues: [{ text: "spoken words", start: 1, end: 2, source: "client" }],
         });
         return {
-          description: "[Video description: caption; transcript[source=client] spoken words]",
+          description,
           durationSeconds: 2,
           framesRequested: 1,
           framesUsed: 1,
@@ -170,6 +172,9 @@ test("reports only validated transcript provenance in guardrail metadata", async
     {}
   );
   assert.equal(result.meta?.transcriptCuesApplied, 1);
+  assert.deepEqual(result.meta?.videoTranscriptDescriptionFingerprints, [
+    `sha256:${createHash("sha256").update(description).digest("hex")}:${description.length}`,
+  ]);
 });
 
 test("converts Responses input using input_text while preserving sibling order", async () => {

tests/unit/kiro-eventstream-split.test.ts

@@ -33,3 +33,30 @@ test("crc32 is deterministic and ByteQueue buffers bytes", async () => {
   const q = new ByteQueue();
   assert.equal(typeof q, "object");
 });
+
+test("malformed Kiro EventStream payload diagnostics never retain upstream content", async () => {
+  const { crc32, parseEventFrame } = await import("../../open-sse/executors/kiro/eventstream.ts");
+  const transcriptSentinel = "PRIVATE_KIRO_UPSTREAM_TRANSCRIPT_SENTINEL";
+  const payload = new TextEncoder().encode(transcriptSentinel);
+  const frame = new Uint8Array(16 + payload.length);
+  const view = new DataView(frame.buffer);
+  view.setUint32(0, frame.length, false);
+  view.setUint32(4, 0, false);
+  view.setUint32(8, crc32(frame.subarray(0, 8)), false);
+  frame.set(payload, 12);
+  view.setUint32(frame.length - 4, 0, false);
+
+  const warnings: string[] = [];
+  const originalWarn = console.warn;
+  console.warn = (...args: unknown[]) => warnings.push(args.map(String).join(" "));
+  try {
+    assert.deepEqual(parseEventFrame(frame)?.payload, { raw: transcriptSentinel });
+  } finally {
+    console.warn = originalWarn;
+  }
+
+  assert.equal(warnings.length, 1);
+  assert.match(warnings[0], /Failed to parse payload/);
+  assert.match(warnings[0], new RegExp(`${payload.byteLength} bytes`));
+  assert.equal(warnings[0].includes(transcriptSentinel), false);
+});

tests/unit/plugins-hooks.test.ts

@@ -159,6 +159,52 @@ test("runOnResponse passes through if no modification", async () => {
   assert.deepEqual(result, { original: true });
 });
 
+test("native plugin-hook error logs omit transcript echoes for a sensitive request", async () => {
+  const sentinels = [
+    "PRIVATE_PLUGIN_REQUEST_TRANSCRIPT_SENTINEL",
+    "PRIVATE_PLUGIN_RESPONSE_TRANSCRIPT_SENTINEL",
+    "PRIVATE_PLUGIN_ERROR_TRANSCRIPT_SENTINEL",
+  ];
+  const retainedLogs: string[] = [];
+  const originalConsoleError = console.error;
+  console.error = (...args: unknown[]) => {
+    retainedLogs.push(args.map(String).join(" "));
+  };
+  try {
+    registerHook("onRequest", "private-request-plugin", (payload: unknown) => {
+      (payload as Record<string, unknown>).videoTranscriptSensitive = false;
+      throw new Error(sentinels[0]);
+    });
+    registerHook("onResponse", "private-response-plugin", (payload: unknown) => {
+      (payload as Record<string, unknown>).videoTranscriptSensitive = false;
+      throw new Error(sentinels[1]);
+    });
+    registerHook("onError", "private-error-plugin", (payload: unknown) => {
+      (payload as Record<string, unknown>).videoTranscriptSensitive = false;
+      throw new Error(sentinels[2]);
+    });
+
+    await emitHookBlocking("onRequest", { videoTranscriptSensitive: true });
+    await runOnResponse(
+      {
+        requestId: "private-request",
+        body: {},
+        model: "test",
+        metadata: {},
+        videoTranscriptSensitive: true,
+      },
+      { ok: true }
+    );
+    await emitHook("onError", { videoTranscriptSensitive: true });
+  } finally {
+    console.error = originalConsoleError;
+  }
+
+  const retained = retainedLogs.join("\n");
+  for (const sentinel of sentinels) assert.equal(retained.includes(sentinel), false);
+  assert.match(retained, /omitted: video transcript/);
+});
+
 // ── runOnError ──
 
 test("runOnError fires emitHook", async () => {

tests/unit/provider-error-detail-lastError.test.ts

@@ -85,8 +85,12 @@ test("extractErrorMessage stays available to toJsonErrorPayload's callers", () =
 test("markAccountUnavailable routes lastError through the helper", () => {
   const src = fs.readFileSync(new URL("../../src/sse/services/auth.ts", import.meta.url), "utf8");
   assert.ok(
-    src.includes("describeUpstreamFailure(errorText)"),
+    src.includes("const retainedErrorText = options.retainedErrorText ?? errorText"),
-    "auth.ts must describe the failure instead of discarding non-string errors"
+    "auth.ts must retain the caller-selected safe error text"
+  );
+  assert.ok(
+    src.includes("describeUpstreamFailure(retainedErrorText)"),
+    "auth.ts must describe the retained failure instead of echoing the raw provider body"
   );
   assert.equal(
     /typeof errorText === "string" \? errorText\.slice\(0, 100\) : "Provider error"/.test(src),

tests/unit/reasoning-cache.test.ts

@@ -182,6 +182,24 @@ describe("Reasoning Replay Cache — Service Layer", () => {
     assert.equal(lookupReasoning("call_capture_2"), "Captured assistant reasoning");
   });
 
+  it("should never cache reasoning echoed by a transcript-sensitive response", () => {
+    clearReasoningCacheAll();
+    const sentinel = "PRIVATE_REASONING_CACHE_TRANSCRIPT_SENTINEL";
+    const cached = cacheReasoningFromAssistantMessage(
+      {
+        role: "assistant",
+        reasoning_content: sentinel,
+        tool_calls: [{ id: "call_private_video_reasoning" }],
+      },
+      "deepseek",
+      "deepseek-reasoner",
+      { videoTranscriptSensitive: true }
+    );
+
+    assert.equal(cached, 0);
+    assert.equal(lookupReasoning("call_private_video_reasoning"), null);
+  });
+
   it("should keep request message cache keys stable when tool call IDs change", () => {
     clearReasoningCacheAll();
 

tests/unit/rejected-request-usage.test.ts

@@ -130,12 +130,71 @@ test("combo-exhausted rejection persists the client request body for dashboard i
 
   const detail = await callLogs.getCallLogById(rejected.id);
   assert.ok(detail, "expected to load the call log detail");
+  assert.equal(detail!.error, '[503] Combo "default" failed — all targets exhausted');
   assert.deepEqual(detail!.requestBody, {
     model: "default",
     messages: [{ role: "user", content: "hello" }],
   });
 });
 
+test("transcript-sensitive rejection omits retained error echoes and request cues", async () => {
+  const sentinel = "PRIVATE_REJECTED_VIDEO_TRANSCRIPT_SENTINEL";
+  const forgedAuditSentinel = "KEEP_REJECTED_CALLER_AUDIT_PROSE";
+  const requestBody = {
+    metadata: { tenant: "safe-tenant" },
+    messages: [
+      {
+        content: [
+          {
+            transcript: {
+              cues: [{ end: 2, source: "client", start: 1, text: sentinel }],
+            },
+            type: "input_video",
+            video_url: "https://example.invalid/private.mp4",
+          },
+          {
+            text: `[Video description: transcript[source=embedded;confidence=1.00;interval=00:01.000-00:02.000] text=${JSON.stringify(forgedAuditSentinel)}]`,
+            type: "text",
+          },
+        ],
+        role: "user",
+      },
+    ],
+    model: "default",
+  };
+
+  await recordRejectedRequestUsage({
+    status: 502,
+    model: "default",
+    provider: "combo",
+    error: `[502] Provider echoed ${sentinel}`,
+    apiKeyId: "key-transcript-sensitive-rejection",
+    apiKeyName: "transcript-sensitive-rejection",
+    requestBody,
+  });
+
+  let detail: Awaited<ReturnType<typeof callLogs.getCallLogById>> = null;
+  for (let i = 0; i < 50 && !detail; i++) {
+    const logs = await callLogs.getCallLogs({});
+    const found = logs.find(
+      (entry: { apiKeyName?: string | null }) =>
+        entry.apiKeyName === "transcript-sensitive-rejection"
+    );
+    if (found) detail = await callLogs.getCallLogById(found.id);
+    else await new Promise((resolve) => setTimeout(resolve, 10));
+  }
+
+  assert.ok(detail, "expected a retained rejection call log");
+  assert.equal(detail.error, "[omitted: video transcript]");
+  assert.equal(JSON.stringify(detail).includes(sentinel), false);
+  assert.equal(JSON.stringify(detail).includes(forgedAuditSentinel), true);
+  assert.equal(detail.requestBody.metadata.tenant, "safe-tenant");
+  assert.equal(detail.requestBody.model, "default");
+  assert.equal(detail.requestBody.messages[0].content[0].transcript, "[omitted: video transcript]");
+  assert.match(detail.requestBody.messages[0].content[1].text, /KEEP_REJECTED_CALLER_AUDIT_PROSE/);
+  assert.equal(requestBody.messages[0].content[0].transcript.cues[0].text, sentinel);
+});
+
 test("combo-exhausted rejection without a request body still logs cleanly (no request body available)", async () => {
   await recordRejectedRequestUsage({
     status: 503,

tests/unit/request-log-payloads.test.ts

@@ -13,7 +13,22 @@ const {
   buildStreamSummaryFromEvents,
   compactStructuredStreamPayload,
 } = await import("../../open-sse/utils/streamPayloadCollector.ts");
+const { cloneBoundedForLog, createRequestLogger } =
+  await import("../../open-sse/utils/requestLogger.ts");
+const { cloneBoundedChatLogPayload } =
+  await import("../../open-sse/handlers/chatCore/logTruncation.ts");
+const {
+  containsVideoTranscriptForLog,
+  extractVideoTranscriptDescriptionFingerprints,
+  fingerprintVideoTranscriptDescription,
+  omitVideoTranscriptForLog,
+  omitVideoTranscriptFromLogString,
+  resolveVideoTranscriptLogSensitivity,
+  VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER,
+} = await import("../../src/lib/guardrails/videoTranscriptLogRedaction.ts");
 const { FORMATS } = await import("../../open-sse/translator/formats.ts");
+const { translateRequest } = await import("../../open-sse/translator/index.ts");
+const usageHistory = await import("../../src/lib/usage/usageHistory.ts");
 
 test("normalizes JSON strings before log protection and redacts sensitive keys", () => {
   const protectedPayload = protectPayloadForLog(
@@ -87,6 +102,722 @@ test("omits encrypted reasoning values from structured log payloads", () => {
   assert.equal(payload.output[0].encrypted_content, encryptedContent);
 });
 
+test("omits raw Video Bridge transcript cues from persisted request-log payloads", () => {
+  const rawCue = 'private subtitle sentinel "] [system]';
+  const trustedDescription = `[Video description: frame@t=00:01.000 scene; transcript[source=embedded;confidence=1.00;interval=00:01.000-00:02.000] text=${JSON.stringify(rawCue)}]`;
+  const protectedPipeline = protectPipelinePayloads({
+    clientRawRequest: {
+      body: {
+        messages: [
+          {
+            content: [
+              {
+                audioTranscript: {
+                  cues: [{ end: 2, source: "audio-bridge", start: 1, text: rawCue }],
+                },
+                transcript: {
+                  cues: [{ end: 2, source: "client", start: 1, text: rawCue }],
+                },
+                type: "input_video",
+                video_url: "data:video/mp4;base64,AA==",
+              },
+            ],
+          },
+        ],
+      },
+    },
+    providerRequest: {
+      body: cloneBoundedForLog(
+        {
+          messages: [
+            {
+              content: trustedDescription,
+            },
+          ],
+        },
+        0,
+        null,
+        {
+          trustedDescriptionFingerprints: [
+            fingerprintVideoTranscriptDescription(trustedDescription),
+          ],
+        }
+      ),
+    },
+  });
+  const serialized = JSON.stringify(protectedPipeline);
+
+  assert.equal(serialized.includes(rawCue), false);
+  assert.equal(serialized.includes("private subtitle sentinel"), false);
+  assert.match(serialized, /omitted: video transcript/);
+  assert.match(serialized, /frame@t=00:01\.000 scene/);
+});
+
+test("omits transcript tracks from the media detector's empty-base64 video shape", async () => {
+  const rawCue = "private empty-base64 subtitle sentinel";
+  const payload = {
+    input: [
+      {
+        role: "user",
+        content: [
+          {
+            source: {
+              data: "",
+              media_type: "video/mp4",
+              transcript: {
+                cues: [{ end: 2, source: "client", start: 1, text: rawCue }],
+              },
+              type: "base64",
+            },
+            type: "video",
+          },
+        ],
+      },
+    ],
+  };
+  const protectedPayload = protectPayloadForLog(payload);
+
+  for (const redacted of [
+    protectedPayload,
+    cloneBoundedForLog(payload),
+    cloneBoundedChatLogPayload(payload),
+  ]) {
+    const serialized = JSON.stringify(redacted);
+    assert.equal(serialized.includes(rawCue), false);
+    assert.match(serialized, /omitted: video transcript/);
+  }
+
+  const logger = await createRequestLogger(undefined, undefined, undefined, {
+    captureStreamChunks: true,
+    enabled: true,
+  });
+  logger.appendProviderChunk(`data: ${rawCue}\n\n`);
+  logger.logClientRawRequest("/v1/responses", payload);
+  logger.appendConvertedChunk(`data: ${rawCue}\n\n`);
+  const pipeline = logger.getPipelinePayloads();
+  assert.equal(JSON.stringify(pipeline).includes(rawCue), false);
+  assert.equal(pipeline?.streamChunks, undefined);
+});
+
+test("omits Video Bridge cues before lossy request-log string truncation", () => {
+  const rawCue = "private boundary subtitle sentinel";
+  const payload = (targetCue: number, visualPadding: number) => {
+    const cueText = (index: number): string =>
+      index === targetCue
+        ? rawCue + "x".repeat(4_096 - rawCue.length)
+        : String(index).padStart(2, "0") + "x".repeat(4_094);
+    const description = `[Video description: ${Array.from(
+      { length: 16 },
+      (_unused, index) =>
+        `transcript[source=client;confidence=1.00;interval=00:${String(index).padStart(2, "0")}.000-00:${String(index + 1).padStart(2, "0")}.000] text=${JSON.stringify(cueText(index))}`
+    ).join("; ")}; frame-tail=${"v".repeat(visualPadding)}]`;
+    return {
+      context: {
+        trustedDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+      },
+      value: { input: [{ content: description, role: "user" }] },
+    };
+  };
+  const requestLogNine = payload(9, 3_507);
+  const requestLogTen = payload(10, 7_710);
+  const chatLogNine = payload(9, 3_522);
+  const alternateChatLogNine = payload(9, 3_598);
+  const boundedPayloads = [
+    cloneBoundedForLog(requestLogNine.value, 0, null, requestLogNine.context),
+    cloneBoundedForLog(requestLogTen.value, 0, null, requestLogTen.context),
+    cloneBoundedChatLogPayload(chatLogNine.value, 0, chatLogNine.context),
+    cloneBoundedChatLogPayload(alternateChatLogNine.value, 0, alternateChatLogNine.context),
+  ];
+
+  for (const bounded of boundedPayloads) {
+    const serialized = JSON.stringify(protectPayloadForLog(bounded));
+    assert.equal(serialized.includes(rawCue), false);
+    assert.match(serialized, /omitted: video transcript/);
+  }
+});
+
+test("suppresses bounded stream-chunk logs for a request carrying Video Bridge cues", async () => {
+  const rawCue = "private streamed subtitle sentinel";
+  const description =
+    `[Video description: transcript[source=embedded;confidence=1.00;` +
+    `interval=00:01.000-00:02.000] text=${JSON.stringify(rawCue)}]`;
+  const logger = await createRequestLogger(undefined, undefined, undefined, {
+    captureStreamChunks: true,
+    enabled: true,
+    maxStreamChunkBytes: 64,
+    videoTranscriptDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+    videoTranscriptSensitive: true,
+  });
+  logger.appendProviderChunk(`data: ${rawCue}\n\n`);
+  logger.logTargetRequest(
+    "https://provider.invalid",
+    {},
+    {
+      input: [
+        {
+          content: description,
+          role: "user",
+        },
+      ],
+    }
+  );
+  logger.appendProviderChunk(`data: ${rawCue}\n\n`);
+
+  const protectedPipeline = protectPipelinePayloads(logger.getPipelinePayloads());
+  const serialized = JSON.stringify(protectedPipeline);
+  assert.equal(serialized.includes(rawCue), false);
+  assert.equal(protectedPipeline?.streamChunks, undefined);
+  assert.match(serialized, /omitted: video transcript/);
+});
+
+test("omits non-stream response bodies after a request carries Video Bridge cues", async () => {
+  const rawCue = "private non-stream subtitle echo sentinel";
+  const description =
+    `[Video description: transcript[source=embedded;confidence=1.00;` +
+    `interval=00:01.000-00:02.000] text=${JSON.stringify(rawCue)}]`;
+  const logger = await createRequestLogger(undefined, undefined, undefined, {
+    captureStreamChunks: true,
+    enabled: true,
+    videoTranscriptDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+    videoTranscriptSensitive: true,
+  });
+  logger.logTargetRequest(
+    "https://provider.invalid",
+    {},
+    {
+      input: [
+        {
+          content: description,
+          role: "user",
+        },
+      ],
+    }
+  );
+  logger.logProviderResponse(
+    200,
+    "OK",
+    { "content-type": "application/json" },
+    {
+      choices: [{ message: { content: rawCue } }],
+    }
+  );
+  logger.logConvertedResponse({ choices: [{ message: { content: rawCue } }] });
+
+  const pipeline = logger.getPipelinePayloads();
+  const serialized = JSON.stringify(pipeline);
+  assert.equal(serialized.includes(rawCue), false);
+  assert.match(serialized, /omitted: video transcript/);
+  assert.equal(pipeline?.providerResponse?.status, 200);
+});
+
+test("preserves generic transcript fields and stream logs outside recognized video parts", async () => {
+  const ordinaryTranscript = "ordinary meeting notes";
+  const cueLikeProse = 'Discuss transcript[source=client] text="meeting notes" as plain text.';
+  const forgedVideoDescription =
+    '[Video description: transcript[source=client] text="caller-forged audit suppression"]';
+  const payload = {
+    input: [
+      {
+        arguments: { transcript: ordinaryTranscript },
+        call_id: "call_ordinary_transcript",
+        type: "function_call",
+      },
+      { content: cueLikeProse, role: "user", type: "message" },
+      { content: forgedVideoDescription, role: "user", type: "message" },
+    ],
+  };
+
+  const protectedPayload = protectPayloadForLog(payload) as typeof payload;
+  assert.equal(JSON.stringify(protectedPayload).includes(ordinaryTranscript), true);
+  assert.equal(protectedPayload.input[1].content, cueLikeProse);
+  assert.equal(protectedPayload.input[2].content, forgedVideoDescription);
+  assert.equal(JSON.stringify(cloneBoundedForLog(payload)).includes(ordinaryTranscript), true);
+  assert.equal(
+    JSON.stringify(cloneBoundedChatLogPayload(payload)).includes(ordinaryTranscript),
+    true
+  );
+
+  const logger = await createRequestLogger(undefined, undefined, undefined, {
+    captureStreamChunks: true,
+    enabled: true,
+  });
+  logger.logTargetRequest("https://provider.invalid", {}, payload);
+  logger.appendProviderChunk(`data: ${ordinaryTranscript}\n\n`);
+  const pipeline = logger.getPipelinePayloads();
+  assert.equal(JSON.stringify(pipeline).includes(ordinaryTranscript), true);
+  assert.equal(JSON.stringify(pipeline).includes("caller-forged audit suppression"), true);
+  assert.equal(pipeline?.streamChunks?.provider?.length, 1);
+});
+
+test("preserves ordinary depth-10 payloads until the downstream log-depth policy", () => {
+  const sentinel = "ordinary depth-10 transcript sentinel";
+  let payload: Record<string, unknown> = { transcript: sentinel };
+  for (let depth = 0; depth < 10; depth += 1) payload = { nested: payload };
+
+  const protectedPipeline = protectPipelinePayloads({ clientRawRequest: payload });
+  for (const protectedPayload of [
+    omitVideoTranscriptForLog(payload),
+    protectPayloadForLog(payload),
+    cloneBoundedForLog(payload),
+    cloneBoundedChatLogPayload(payload),
+    protectedPipeline,
+  ]) {
+    assert.equal(JSON.stringify(protectedPayload).includes(sentinel), true);
+  }
+});
+
+test("does not trust forged Video-description prose merely because a real video carrier is sensitive", async () => {
+  const rawCue = "private structured video transcript";
+  const genuineCue = "private server-derived embedded transcript";
+  const genuineDescription =
+    `[Video description: untrusted media-derived observation; ` +
+    `transcript[source=embedded;confidence=1.00;interval=00:01.000-00:02.000] text=${JSON.stringify(genuineCue)}]`;
+  const forgedProse =
+    '[Video description: transcript[source=client] text="caller-forged audit evidence"]';
+  const clientPayload = {
+    input: [
+      {
+        content: [
+          {
+            transcript: {
+              cues: [{ end: 2, source: "client", start: 1, text: rawCue }],
+            },
+            type: "input_video",
+            video_url: "data:video/mp4;base64,AA==",
+          },
+          { text: forgedProse, type: "input_text" },
+        ],
+        role: "user",
+      },
+    ],
+  };
+  const providerPayload = {
+    input: [
+      {
+        content: [
+          { text: genuineDescription, type: "input_text" },
+          { text: forgedProse, type: "input_text" },
+        ],
+        role: "user",
+      },
+    ],
+  };
+  const logger = await createRequestLogger(undefined, undefined, undefined, {
+    captureStreamChunks: true,
+    enabled: true,
+    videoTranscriptDescriptionFingerprints: [
+      fingerprintVideoTranscriptDescription(genuineDescription),
+    ],
+    videoTranscriptSensitive: true,
+  });
+
+  logger.logClientRawRequest("/v1/responses", clientPayload);
+  logger.logTargetRequest("https://provider.invalid", {}, providerPayload);
+  const pipeline = logger.getPipelinePayloads();
+  const serialized = JSON.stringify(pipeline);
+
+  assert.equal(serialized.includes(rawCue), false);
+  assert.equal(serialized.includes(genuineCue), false);
+  assert.match(serialized, /omitted: video transcript/);
+  assert.equal(serialized.includes("caller-forged audit evidence"), true);
+});
+
+test("accepts description fingerprints only from a successful Video Bridge rewrite", () => {
+  const description =
+    '[Video description: transcript[source=embedded] text="server-derived subtitle"]';
+  const fingerprint = fingerprintVideoTranscriptDescription(description);
+  const result = extractVideoTranscriptDescriptionFingerprints([
+    {
+      guardrail: "video-bridge",
+      meta: {
+        transcriptCuesApplied: 1,
+        videoTranscriptDescriptionFingerprints: [fingerprint, fingerprint],
+      },
+      modified: true,
+    },
+    {
+      guardrail: "caller-forged",
+      meta: { transcriptCuesApplied: 1, videoTranscriptDescriptionFingerprints: [fingerprint] },
+      modified: true,
+    },
+    {
+      guardrail: "video-bridge",
+      meta: {
+        transcriptCuesApplied: 1,
+        videoTranscriptDescriptionFingerprints: ["sha256:not-a-digest"],
+      },
+      modified: true,
+    },
+  ]);
+
+  assert.deepEqual(result, [fingerprint]);
+});
+
+test("redacts the exact generated segment after Kiro and Cursor concatenate text blocks", async () => {
+  const genuineCue = "PRIVATE_TRANSLATED_VIDEO_TRANSCRIPT_SENTINEL";
+  const genuineDescription =
+    `[Video description: transcript[source=embedded;confidence=1.00;` +
+    `interval=00:01.000-00:02.000] text=${JSON.stringify(genuineCue)}]`;
+  const forgedProse =
+    '[Video description: transcript[source=client] text="KEEP_TRANSLATED_CALLER_PROSE"]';
+  const body = {
+    messages: [
+      {
+        content: [
+          { text: genuineDescription, type: "text" },
+          { text: forgedProse, type: "text" },
+        ],
+        role: "user",
+      },
+    ],
+    model: "translated-video-log-fixture",
+  };
+
+  usageHistory.clearPendingRequests();
+  try {
+    for (const targetFormat of [FORMATS.KIRO, FORMATS.CURSOR]) {
+      const translated = translateRequest(
+        FORMATS.OPENAI,
+        targetFormat,
+        "translated-video-log-fixture",
+        body,
+        false
+      );
+      const videoTranscriptDescriptionFingerprints = [
+        fingerprintVideoTranscriptDescription(genuineDescription),
+      ];
+      const logger = await createRequestLogger(FORMATS.OPENAI, targetFormat, undefined, {
+        enabled: true,
+        videoTranscriptDescriptionFingerprints,
+        videoTranscriptSensitive: true,
+      });
+      logger.logTargetRequest("https://provider.invalid", {}, translated);
+      const serialized = JSON.stringify(logger.getPipelinePayloads());
+
+      assert.equal(serialized.includes(genuineCue), false, targetFormat);
+      assert.match(serialized, /omitted: video transcript/, targetFormat);
+      assert.equal(serialized.includes("KEEP_TRANSLATED_CALLER_PROSE"), true, targetFormat);
+
+      const requestId = usageHistory.trackPendingRequest(
+        "translated-video-log-fixture",
+        targetFormat,
+        `connection-${targetFormat}`,
+        true,
+        {
+          providerRequest: translated,
+          videoTranscriptDescriptionFingerprints,
+          videoTranscriptSensitive: true,
+        }
+      );
+      const pending = JSON.stringify(usageHistory.getPendingById().get(requestId!));
+      assert.equal(pending.includes(genuineCue), false, targetFormat);
+      assert.match(pending, /omitted: video transcript/, targetFormat);
+      assert.equal(pending.includes("KEEP_TRANSLATED_CALLER_PROSE"), true, targetFormat);
+    }
+  } finally {
+    usageHistory.clearPendingRequests();
+  }
+});
+
+test("fails closed after a bounded number of forged description-prefix candidates", () => {
+  const genuineDescription =
+    '[Video description: transcript[source=embedded] text="bounded genuine subtitle"]';
+  const prefixFlood = Array.from(
+    { length: 129 },
+    (_unused, index) => `[Video description: forged-prefix-${index}]`
+  ).join("");
+  const omitted = omitVideoTranscriptForLog(
+    { content: prefixFlood + genuineDescription },
+    {
+      trustedDescriptionFingerprints: [fingerprintVideoTranscriptDescription(genuineDescription)],
+    }
+  ) as { content: string };
+
+  assert.equal(omitted.content, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER);
+});
+
+test("fails closed after a bounded number of trusted-description candidate hashes", () => {
+  const genuineDescription =
+    '[Video description: transcript[source=embedded] text="bounded hash subtitle"]';
+  const forgedPrefixes = Array.from(
+    { length: 9 },
+    (_unused, index) => `[Video description: candidate-${index}]`
+  ).join("");
+  const trustedDescriptionFingerprints = [
+    ...Array.from({ length: 63 }, (_unused, index) =>
+      fingerprintVideoTranscriptDescription(`[Video description: identity-${index}]`)
+    ),
+    fingerprintVideoTranscriptDescription(genuineDescription),
+  ];
+  const omitted = omitVideoTranscriptForLog(
+    { content: forgedPrefixes + genuineDescription },
+    { trustedDescriptionFingerprints }
+  ) as { content: string };
+
+  assert.equal(omitted.content, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER);
+});
+
+test("trusted malformed or over-budget serialized cues fail closed", () => {
+  const privateTranscript = "PRIVATE_MALFORMED_SERIALIZED_CUE_SENTINEL";
+  const malformedDescriptions = [
+    `[Video description: transcript[source=embedded${";metadata=x".repeat(100)}] text=${JSON.stringify(privateTranscript)}]`,
+    `[Video description: transcript[source=embedded text=${JSON.stringify(privateTranscript)}]`,
+    `[Video description: transcript[source=embedded]text=${JSON.stringify(privateTranscript)}]`,
+    `[Video description: transcript[source=embedded] payload=${JSON.stringify(privateTranscript)}]`,
+    `[Video description: transcript[source=embedded] text=${JSON.stringify("safe cue")}; transcript[source=audio-bridge${";metadata=x".repeat(100)}] text=${JSON.stringify(privateTranscript)}]`,
+  ];
+
+  for (const description of malformedDescriptions) {
+    const context = {
+      trustedDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+    };
+    const direct = omitVideoTranscriptFromLogString(description, context);
+    const structured = omitVideoTranscriptForLog({ content: description }, context) as {
+      content: string;
+    };
+
+    assert.equal(direct, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER, description);
+    assert.equal(structured.content, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER, description);
+    assert.equal(direct.includes(privateTranscript), false, description);
+    assert.equal(structured.content.includes(privateTranscript), false, description);
+  }
+});
+
+test("bounds cyclic objects and fails closed when the transcript-security depth is exceeded", () => {
+  const cyclic: Record<string, unknown> = { transcript: "ordinary cyclic notes" };
+  cyclic.self = cyclic;
+
+  let deep: Record<string, unknown> = { transcript: "ordinary deeply nested notes" };
+  for (let index = 0; index < 20_000; index += 1) deep = { child: deep };
+
+  assert.doesNotThrow(() => containsVideoTranscriptForLog(cyclic));
+  assert.doesNotThrow(() => containsVideoTranscriptForLog(deep));
+  assert.equal(containsVideoTranscriptForLog(cyclic), false);
+  assert.equal(containsVideoTranscriptForLog(deep), true);
+
+  const omittedCycle = omitVideoTranscriptForLog(cyclic);
+  const omittedDeep = omitVideoTranscriptForLog(deep);
+  assert.doesNotThrow(() => JSON.stringify(omittedCycle));
+  assert.equal(omittedDeep, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER);
+});
+
+test("fails closed when enumerable getters or proxies throw during transcript inspection", () => {
+  const throwingGetter: Record<string, unknown> = {};
+  Object.defineProperty(throwingGetter, "nested", {
+    enumerable: true,
+    get() {
+      throw new Error("PRIVATE_GETTER_TRANSCRIPT_SENTINEL");
+    },
+  });
+  const throwingProxy = new Proxy<Record<string, unknown>>(
+    {},
+    {
+      ownKeys() {
+        throw new Error("PRIVATE_PROXY_TRANSCRIPT_SENTINEL");
+      },
+    }
+  );
+
+  for (const payload of [throwingGetter, throwingProxy]) {
+    assert.equal(containsVideoTranscriptForLog(payload), true);
+    assert.equal(omitVideoTranscriptForLog(payload), VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER);
+  }
+});
+
+test("fails closed when an enumerable function can replace the retained JSON representation", () => {
+  const privateTranscript = "PRIVATE_TOJSON_VIDEO_TRANSCRIPT_SENTINEL";
+  const payload = {
+    safe: "retained diagnostic",
+    toJSON() {
+      return {
+        transcript: privateTranscript,
+        type: "input_video",
+        video_url: "data:video/mp4;base64,AA==",
+      };
+    },
+  };
+
+  assert.equal(containsVideoTranscriptForLog(payload), true);
+  const omitted = omitVideoTranscriptForLog(payload) as Record<string, unknown>;
+  const serialized = JSON.stringify(omitted);
+
+  assert.equal(serialized.includes(privateTranscript), false);
+  assert.equal(omitted.safe, "retained diagnostic");
+  assert.notEqual(typeof omitted.toJSON, "function");
+});
+
+test("treats explicit malformed and nested video carriers as transcript-sensitive", () => {
+  const privateTranscript = "PRIVATE_MALFORMED_VIDEO_TRANSCRIPT_SENTINEL";
+  const payloads = [
+    {
+      transcript: privateTranscript,
+      type: "input_video",
+    },
+    {
+      transcript: privateTranscript,
+      type: "video_url",
+      video_url: "",
+    },
+    {
+      input_video: {
+        transcript: privateTranscript,
+        url: "data:video/mp4;base64,AA==",
+      },
+      type: "input_video",
+    },
+    {
+      input_video: {
+        url: { transcript: privateTranscript },
+      },
+      type: "input_video",
+    },
+    {
+      source: { transcript: privateTranscript },
+      type: "video_source",
+    },
+    {
+      source: {
+        media_type: "video/mp4",
+        transcript: privateTranscript,
+      },
+      type: "video",
+    },
+  ];
+
+  for (const payload of payloads) {
+    assert.equal(containsVideoTranscriptForLog(payload), true, JSON.stringify(payload));
+    const serialized = JSON.stringify(omitVideoTranscriptForLog(payload));
+    assert.equal(serialized.includes(privateTranscript), false, serialized);
+    assert.match(serialized, /omitted: video transcript/);
+  }
+});
+
+test("keeps raw request sensitivity after a guardrail removes the original carrier", () => {
+  const rawRequestBody = {
+    messages: [
+      {
+        content: [
+          {
+            transcript: "PRIVATE_RAW_REQUEST_TRANSCRIPT_SENTINEL",
+            type: "input_video",
+            video_url: "data:video/mp4;base64,AA==",
+          },
+        ],
+        role: "user",
+      },
+    ],
+  };
+  const processedBody = {
+    messages: [{ content: "guardrail replaced the media", role: "user" }],
+  };
+
+  assert.equal(resolveVideoTranscriptLogSensitivity({ processedBody, rawRequestBody }), true);
+  assert.equal(
+    resolveVideoTranscriptLogSensitivity({
+      processedBody: { metadata: { transcript: "ordinary audit label" } },
+      rawRequestBody: { metadata: { transcript: "ordinary caller label" } },
+    }),
+    false
+  );
+});
+
+test("fails closed at the aggregate traversal budget before a tail video transcript can leak", () => {
+  const privateTranscript = "private over-budget video transcript sentinel";
+  const payload: unknown[] = Array.from(
+    { length: 10_001 },
+    (_unused, index) => `ordinary entry ${index}`
+  );
+
+  // The detector must treat a bounded security scan as unknown/sensitive even without a cue.
+  assert.equal(containsVideoTranscriptForLog(payload), true);
+
+  payload[payload.length - 1] = {
+    transcript: privateTranscript,
+    type: "input_video",
+    video_url: "data:video/mp4;base64,AA==",
+  };
+  const omitted = omitVideoTranscriptForLog(payload);
+  assert.equal(omitted, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER);
+  assert.equal(JSON.stringify(omitted).includes(privateTranscript), false);
+});
+
+test("detects and omits an input-video transcript below nine ordinary objects", () => {
+  const privateTranscript = "private depth-9 video transcript sentinel";
+  let payload: Record<string, unknown> = {
+    transcript: privateTranscript,
+    type: "input_video",
+    video_url: { url: "data:video/mp4;base64,AA==" },
+  };
+  for (let depth = 0; depth < 9; depth += 1) payload = { nested: payload };
+
+  assert.equal(containsVideoTranscriptForLog(payload), true);
+  const omitted = JSON.stringify(omitVideoTranscriptForLog(payload));
+  assert.equal(omitted.includes(privateTranscript), false);
+  assert.match(omitted, /omitted: video transcript/);
+});
+
+test("redacts only direct transcript carriers within a recognized video part", () => {
+  const directCue = "private direct video subtitle";
+  const sourceCue = "private source video subtitle";
+  const unrelatedNestedTranscript = "ordinary nested metadata transcript";
+  const payload = {
+    type: "video",
+    source: {
+      data: "AA==",
+      media_type: "video/mp4",
+      transcript: sourceCue,
+      metadata: { transcript: unrelatedNestedTranscript },
+      type: "base64",
+    },
+    transcript: directCue,
+    metadata: { transcript: unrelatedNestedTranscript },
+  };
+
+  const omitted = omitVideoTranscriptForLog(payload) as typeof payload;
+  assert.equal(omitted.transcript, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER);
+  assert.equal(omitted.metadata.transcript, unrelatedNestedTranscript);
+  assert.equal(omitted.source.transcript, VIDEO_TRANSCRIPT_LOG_OMISSION_MARKER);
+  assert.equal(omitted.source.metadata.transcript, unrelatedNestedTranscript);
+});
+
+test("suppresses active stream chunks even when persisted pipeline logging is disabled", async () => {
+  const rawCue = "private active-log subtitle sentinel";
+  const model = "video-log-redaction-model";
+  const provider = "video-log-redaction-provider";
+  const connectionId = "video-log-redaction-connection";
+  usageHistory.clearPendingRequests();
+  try {
+    const requestId = usageHistory.trackPendingRequest(model, provider, connectionId, true);
+    const logger = await createRequestLogger(undefined, undefined, model, {
+      captureStreamChunks: true,
+      connectionId,
+      enabled: false,
+      model,
+      provider,
+      requestId,
+    });
+    logger.appendProviderChunk(`data: ${rawCue}\n\n`);
+    logger.logClientRawRequest("/v1/responses", {
+      input: [
+        {
+          audioTranscript: {
+            cues: [{ end: 2, source: "audio-bridge", start: 1, text: rawCue }],
+          },
+          type: "input_video",
+          video_url: "data:video/mp4;base64,AA==",
+        },
+      ],
+    });
+    logger.appendOpenAIChunk(`data: ${rawCue}\n\n`);
+    logger.appendConvertedChunk(`data: ${rawCue}\n\n`);
+
+    const chunks = usageHistory.getPendingById().get(requestId)?.streamChunks;
+    assert.deepEqual(chunks, { client: [], openai: [], provider: [] });
+  } finally {
+    usageHistory.clearPendingRequests();
+  }
+});
+
 test("omits encrypted reasoning split across captured SSE chunks", () => {
   const encryptedContent = "opaque-replay-state".repeat(128);
   const protectedPipeline = protectPipelinePayloads({

tests/unit/request-logger-endpoints.test.ts

@@ -8,7 +8,10 @@ const TEST_DATA_DIR = fs.mkdtempSync(path.join(os.tmpdir(), "omniroute-reqlogger
 process.env.DATA_DIR = TEST_DATA_DIR;
 
 const core = await import("../../src/lib/db/core.ts");
+const { fingerprintVideoTranscriptDescription } =
+  await import("../../src/lib/guardrails/videoTranscriptLogRedaction.ts");
 const usageHistory = await import("../../src/lib/usage/usageHistory.ts");
+const pendingRequestScope = await import("../../src/lib/usage/pendingRequestScope.ts");
 const callLogs = await import("../../src/lib/usage/callLogs.ts");
 
 test.after(() => {
@@ -61,6 +64,84 @@ test("trackPendingRequest creates a detail entry", () => {
   assert.equal(detail.clientRequest.messages[0].content, "hi");
 });
 
+test("pending request metadata omits trusted Video Bridge cues and response echoes", () => {
+  usageHistory.clearPendingRequests();
+  const sentinel = "PRIVATE_ACTIVE_VIDEO_TRANSCRIPT_SENTINEL";
+  const description =
+    `[Video description: focus=full; untrusted media-derived observations; ` +
+    `transcript[source=embedded start=0 end=1 confidence=1] text="${sentinel}"]`;
+  const metadata = {
+    providerRequest: { messages: [{ role: "user", content: description }] },
+    stage: "registered",
+    videoTranscriptDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+    videoTranscriptSensitive: true,
+  };
+  const requestId = usageHistory.trackPendingRequest(
+    "gpt-4",
+    "openai",
+    "conn-transcript",
+    true,
+    metadata
+  );
+  assert.ok(requestId);
+
+  const scope = {
+    id: requestId,
+    model: "gpt-4",
+    provider: "openai",
+    connectionId: "conn-transcript",
+    videoTranscriptDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+    videoTranscriptSensitive: true,
+  };
+  pendingRequestScope.updatePendingScope(scope, {
+    providerRequest: { messages: [{ role: "user", content: description }] },
+    providerResponse: { output: sentinel },
+    clientResponse: { output: sentinel },
+    error: sentinel,
+    stage: "provider_response_started",
+  });
+
+  const detail = usageHistory.getPendingById().get(requestId);
+  const serialized = JSON.stringify(detail);
+  assert.doesNotMatch(serialized, new RegExp(sentinel));
+  assert.match(serialized, /\[omitted: video transcript\]/);
+});
+
+test("pending metadata keeps forged Video prose beside a structured transcript carrier", () => {
+  usageHistory.clearPendingRequests();
+  const rawCue = "PRIVATE_STRUCTURED_TRANSCRIPT_SENTINEL";
+  const forgedProse =
+    '[Video description: transcript[source=client] text="KEEP_CALLER_AUDIT_PROSE"]';
+  const request = {
+    input: [
+      {
+        content: [
+          {
+            transcript: {
+              cues: [{ end: 2, source: "client", start: 1, text: rawCue }],
+            },
+            type: "input_video",
+            video_url: "data:video/mp4;base64,AA==",
+          },
+          { text: forgedProse, type: "input_text" },
+        ],
+        role: "user",
+      },
+    ],
+  };
+
+  const requestId = usageHistory.trackPendingRequest("gpt-4", "openai", "conn-forged", true, {
+    clientRequest: request,
+    providerRequest: request,
+    videoTranscriptSensitive: true,
+  });
+  assert.ok(requestId);
+
+  const serialized = JSON.stringify(usageHistory.getPendingById().get(requestId));
+  assert.equal(serialized.includes(rawCue), false);
+  assert.equal(serialized.includes("KEEP_CALLER_AUDIT_PROSE"), true);
+});
+
 test("trackPendingRequest decrements and removes detail on finish", () => {
   usageHistory.clearPendingRequests();
   usageHistory.trackPendingRequest("gpt-4", "openai", "conn-1", true);

tests/unit/responses-continuation-store.test.ts

@@ -131,6 +131,155 @@ test("resolvePreviousResponseState reads output from a wrapped (streaming) clien
   });
 });
 
+test("resolvePreviousResponseState continues from a safely redacted Video Bridge transcript", () => {
+  insertCallLog({
+    id: "log-video-redacted",
+    responseId: "resp_video_redacted",
+    apiKeyId: "key-1",
+    detailState: "ready",
+    artifactRelPath: "2026-01-01/log-video-redacted.json",
+  });
+  const request = {
+    body: {
+      input: [
+        {
+          role: "user",
+          content:
+            '[Video description: transcript[source=embedded] text="[omitted: video transcript]"]',
+        },
+      ],
+    },
+  };
+  writeArtifact("2026-01-01/log-video-redacted.json", {
+    _omnirouteVideoTranscriptRedacted: true,
+    clientRawRequest: request,
+    providerRequest: request,
+    clientResponse: {
+      id: "resp_video_redacted",
+      output: [{ type: "message", role: "assistant", content: "summary" }],
+    },
+  });
+
+  assert.deepEqual(store.resolvePreviousResponseState("resp_video_redacted", "key-1"), {
+    input: request.body.input,
+    output: [{ type: "message", role: "assistant", content: "summary" }],
+  });
+});
+
+test("resolvePreviousResponseState continues when the prior output was safely redacted", () => {
+  insertCallLog({
+    id: "log-video-output-redacted",
+    responseId: "resp_video_output_redacted",
+    apiKeyId: "key-1",
+    detailState: "ready",
+    artifactRelPath: "2026-01-01/log-video-output-redacted.json",
+  });
+  const request = {
+    body: { input: [{ role: "user", content: "summarize the prior result" }] },
+  };
+  writeArtifact("2026-01-01/log-video-output-redacted.json", {
+    _omnirouteVideoTranscriptRedacted: true,
+    clientRawRequest: request,
+    providerRequest: request,
+    clientResponse: {
+      id: "resp_video_output_redacted",
+      output: [
+        {
+          type: "message",
+          role: "assistant",
+          content: '[Video description: text="[omitted: video transcript]"]',
+        },
+      ],
+    },
+  });
+
+  assert.deepEqual(store.resolvePreviousResponseState("resp_video_output_redacted", "key-1"), {
+    input: request.body.input,
+    output: [
+      {
+        type: "message",
+        role: "assistant",
+        content: '[Video description: text="[omitted: video transcript]"]',
+      },
+    ],
+  });
+});
+
+test("resolvePreviousResponseState continues from an empty-base64 transcript omission", () => {
+  insertCallLog({
+    id: "log-empty-base64-video-redacted",
+    responseId: "resp_empty_base64_video_redacted",
+    apiKeyId: "key-1",
+    detailState: "ready",
+    artifactRelPath: "2026-01-01/log-empty-base64-video-redacted.json",
+  });
+  const request = {
+    body: {
+      input: [
+        {
+          type: "video",
+          source: {
+            data: "",
+            media_type: "video/mp4",
+            transcript: "[omitted: video transcript]",
+            type: "base64",
+          },
+        },
+      ],
+    },
+  };
+  writeArtifact("2026-01-01/log-empty-base64-video-redacted.json", {
+    _omnirouteVideoTranscriptRedacted: true,
+    clientRawRequest: request,
+    providerRequest: request,
+    clientResponse: {
+      id: "resp_empty_base64_video_redacted",
+      output: [{ type: "message", role: "assistant", content: "summary" }],
+    },
+  });
+
+  assert.deepEqual(
+    store.resolvePreviousResponseState("resp_empty_base64_video_redacted", "key-1"),
+    {
+      input: request.body.input,
+      output: [{ type: "message", role: "assistant", content: "summary" }],
+    }
+  );
+});
+
+test("resolvePreviousResponseState preserves ordinary transcript fields and marker-like prose", () => {
+  insertCallLog({
+    id: "log-ordinary-transcript",
+    responseId: "resp_ordinary_transcript",
+    apiKeyId: "key-1",
+    detailState: "ready",
+    artifactRelPath: "2026-01-01/log-ordinary-transcript.json",
+  });
+  const input = [
+    {
+      arguments: { transcript: "ordinary meeting notes" },
+      type: "function_call",
+    },
+    {
+      content:
+        '[Video description: caller prose with text="[omitted: video transcript]"] without a video.',
+      role: "user",
+      type: "message",
+    },
+  ];
+  const output = [{ type: "message", role: "assistant", content: "ordinary reply" }];
+  writeArtifact("2026-01-01/log-ordinary-transcript.json", {
+    clientRawRequest: { body: { input } },
+    providerRequest: { body: { input } },
+    clientResponse: { id: "resp_ordinary_transcript", output },
+  });
+
+  assert.deepEqual(store.resolvePreviousResponseState("resp_ordinary_transcript", "key-1"), {
+    input,
+    output,
+  });
+});
+
 test("resolvePreviousResponseState returns null for an unknown response id", () => {
   const result = store.resolvePreviousResponseState("resp_does_not_exist", "key-1");
   assert.equal(result, null);

tests/unit/stream-finalization-video-retention.test.ts

@@ -0,0 +1,61 @@
+import assert from "node:assert/strict";
+import fs from "node:fs";
+import os from "node:os";
+import path from "node:path";
+import test from "node:test";
+
+const TEST_DATA_DIR = fs.mkdtempSync(path.join(os.tmpdir(), "omniroute-stream-retention-"));
+process.env.DATA_DIR = TEST_DATA_DIR;
+
+const core = await import("../../src/lib/db/core.ts");
+const usageHistory = await import("../../src/lib/usage/usageHistory.ts");
+const { finalizeStreamRequestLog } =
+  await import("../../open-sse/utils/streamFailureFinalization.ts");
+
+test.after(() => {
+  usageHistory.clearPendingRequests();
+  core.resetDbInstance();
+  fs.rmSync(TEST_DATA_DIR, { force: true, recursive: true });
+});
+
+test("stream finalization omits transcript-sensitive response echoes", () => {
+  usageHistory.clearPendingRequests();
+  const sentinel = "PRIVATE_STREAM_FINALIZATION_TRANSCRIPT_SENTINEL";
+  const requestId = usageHistory.trackPendingRequest(
+    "video-model",
+    "video-provider",
+    "video-connection",
+    true,
+    { videoTranscriptSensitive: true }
+  );
+  assert.ok(requestId);
+
+  finalizeStreamRequestLog({
+    pendingRequestId: requestId,
+    model: "video-model",
+    provider: "video-provider",
+    connectionId: "video-connection",
+    providerResponse: { echo: sentinel },
+    clientResponse: { echo: sentinel },
+    status: 502,
+    error: sentinel,
+    videoTranscriptSensitive: true,
+  });
+
+  const retained = usageHistory.getCompletedDetails().get(requestId);
+  assert.ok(retained);
+  const serialized = JSON.stringify(retained);
+  assert.equal(serialized.includes(sentinel), false);
+  assert.match(serialized, /omitted: video transcript/);
+});
+
+test("chatCore propagates the request sensitivity bit into stream finalization", () => {
+  const source = fs.readFileSync("open-sse/handlers/chatCore.ts", "utf8");
+  const callStart = source.indexOf("streamFailure.finalizeStreamRequestLog({");
+  assert.notEqual(callStart, -1, "stream finalization call must exist");
+  const callEnd = source.indexOf("\n    });", callStart);
+  assert.notEqual(callEnd, -1, "stream finalization call must be bounded");
+  const callSource = source.slice(callStart, callEnd);
+
+  assert.match(callSource, /\n\s+videoTranscriptSensitive,\s*$/m);
+});

tests/unit/stream-handler-catch-logging-8143.test.ts

@@ -51,9 +51,14 @@ test("pipeWithDisconnect stall watchdog logs instead of silently swallowing a th
   };
 
   try {
-    const stream = pipeWithDisconnect(new Response(source), new TransformStream(), streamController, {
+    const stream = pipeWithDisconnect(
-      stallTimeoutMs: 40,
+      new Response(source),
-    });
+      new TransformStream(),
+      streamController,
+      {
+        stallTimeoutMs: 40,
+      }
+    );
     await readStreamText(stream);
   } finally {
     console.debug = originalDebug;
@@ -67,3 +72,45 @@ test("pipeWithDisconnect stall watchdog logs instead of silently swallowing a th
     "a throwing handleError during the stall watchdog must be logged via console.debug, not swallowed"
   );
 });
+
+test("pipeWithDisconnect omits transcript-sensitive stall watchdog diagnostics", async () => {
+  const privateCue = "PRIVATE_VIDEO_CUE_stall_watchdog_4c91";
+  const source = new ReadableStream({
+    start(controller) {
+      controller.enqueue(encoder.encode("x"));
+    },
+    cancel() {},
+  });
+  const streamController = {
+    isConnected: () => true,
+    handleError() {
+      throw new Error(privateCue);
+    },
+    handleComplete() {},
+    abort() {},
+  };
+  const debugCalls = [];
+  const originalDebug = console.debug;
+  console.debug = (...args) => {
+    debugCalls.push(args);
+  };
+
+  try {
+    const stream = pipeWithDisconnect(
+      new Response(source),
+      new TransformStream(),
+      streamController,
+      {
+        redactStreamDiagnosticsForLog: true,
+        stallTimeoutMs: 40,
+      }
+    );
+    await readStreamText(stream);
+  } finally {
+    console.debug = originalDebug;
+  }
+
+  const retainedDiagnostics = debugCalls.flat().map(String).join(" ");
+  assert.equal(retainedDiagnostics.includes(privateCue), false);
+  assert.equal(retainedDiagnostics.includes("[omitted: video transcript]"), true);
+});

tests/unit/stream-handler.test.ts

@@ -829,3 +829,38 @@ test("pipeWithDisconnect stall watchdog does not fire after normal stream comple
   assert.equal(text, "ok");
   assert.equal(onErrorCalled, false, "stall watchdog must be cleared on stream completion");
 });
+
+test("createStreamController redacts sensitive diagnostics without changing onError", () => {
+  const privateCue = "PRIVATE_VIDEO_CUE_controller_7f38";
+  const originalLog = console.log;
+  const originalDebug = console.debug;
+  const calls: unknown[][] = [];
+  let callbackMessage: string | null = null;
+
+  console.log = (...args: unknown[]) => {
+    calls.push(args);
+  };
+  console.debug = (...args: unknown[]) => {
+    calls.push(args);
+  };
+
+  try {
+    const controller = createStreamController({
+      redactStreamDiagnosticsForLog: true,
+      onError(event) {
+        callbackMessage = event.message;
+        throw new Error(`${privateCue}: callback failure`);
+      },
+    });
+
+    controller.handleError(new Error(privateCue));
+  } finally {
+    console.log = originalLog;
+    console.debug = originalDebug;
+  }
+
+  const retainedDiagnostics = calls.flat().map(String).join(" ");
+  assert.equal(callbackMessage, privateCue, "fallback classification must receive the real error");
+  assert.equal(retainedDiagnostics.includes(privateCue), false);
+  assert.equal(retainedDiagnostics.includes("[omitted: video transcript]"), true);
+});

tests/unit/stream-onfailure-callback-logging.test.ts

@@ -80,6 +80,11 @@ test("#8142 onFailure throwing does not crash the stream and is logged", async (
     loggedOnFailureThrow,
     "a console.debug call referencing onFailure must be emitted when the callback throws"
   );
+  assert.match(
+    debugCalls.flat().map(String).join(" "),
+    /boom from consumer onFailure handler/,
+    "ordinary stream diagnostics must retain their useful error message"
+  );
 });
 
 test("#8142 regression: onFailure returning normally logs nothing and behaves identically", async () => {
@@ -117,3 +122,112 @@ test("#8142 regression: onFailure returning normally logs nothing and behaves id
     "the happy path (no throw) must not emit the onFailure-throw debug log — behavior-free regression guard"
   );
 });
+
+test("sensitive onFailure diagnostics omit embedded transcript text", async () => {
+  const privateCue = "PRIVATE_VIDEO_CUE_onFailure_2e91";
+  const originalDebug = console.debug;
+  const debugCalls: unknown[][] = [];
+  console.debug = (...args: unknown[]) => {
+    debugCalls.push(args);
+  };
+
+  try {
+    await assert.rejects(
+      readTransformed([responseFailedChunk("upstream failed")], {
+        mode: "passthrough",
+        sourceFormat: FORMATS.OPENAI,
+        provider: "openai",
+        model: "gpt-test",
+        body: { messages: [{ role: "user", content: "hello" }] },
+        redactStreamDiagnosticsForLog: true,
+        onFailure() {
+          throw new Error(privateCue);
+        },
+      }),
+      /upstream failed/i
+    );
+  } finally {
+    console.debug = originalDebug;
+  }
+
+  const retainedDiagnostics = debugCalls.flat().map(String).join(" ");
+  assert.equal(retainedDiagnostics.includes(privateCue), false);
+  assert.equal(retainedDiagnostics.includes("[omitted: video transcript]"), true);
+});
+
+test("sensitive onComplete diagnostics omit embedded transcript text", async () => {
+  const privateCue = "PRIVATE_VIDEO_CUE_onComplete_8d42";
+  const originalDebug = console.debug;
+  const debugCalls: unknown[][] = [];
+  console.debug = (...args: unknown[]) => {
+    debugCalls.push(args);
+  };
+
+  try {
+    await readTransformed(
+      [
+        `data: ${JSON.stringify({
+          id: "chatcmpl-sensitive-log",
+          choices: [{ index: 0, delta: { content: "ok" }, finish_reason: "stop" }],
+        })}\n\n`,
+        "data: [DONE]\n\n",
+      ],
+      {
+        mode: "passthrough",
+        sourceFormat: FORMATS.OPENAI,
+        provider: "openai",
+        model: "gpt-test",
+        body: { messages: [{ role: "user", content: "hello" }] },
+        redactStreamDiagnosticsForLog: true,
+        onComplete() {
+          throw new Error(privateCue);
+        },
+      }
+    );
+  } finally {
+    console.debug = originalDebug;
+  }
+
+  const retainedDiagnostics = debugCalls.flat().map(String).join(" ");
+  assert.equal(retainedDiagnostics.includes(privateCue), false);
+  assert.equal(retainedDiagnostics.includes("[omitted: video transcript]"), true);
+});
+
+test("sensitive flush diagnostics omit embedded transcript text", async () => {
+  const privateCue = "PRIVATE_VIDEO_CUE_flush_f601";
+  const originalLog = console.log;
+  const logCalls: unknown[][] = [];
+  console.log = (...args: unknown[]) => {
+    logCalls.push(args);
+  };
+
+  try {
+    await readTransformed(
+      [
+        `data: ${JSON.stringify({
+          id: "chatcmpl-sensitive-flush",
+          choices: [{ index: 0, delta: { content: "tail" }, finish_reason: "stop" }],
+        })}`,
+      ],
+      {
+        mode: "passthrough",
+        sourceFormat: FORMATS.OPENAI,
+        provider: "openai",
+        model: "gpt-test",
+        body: { messages: [{ role: "user", content: "hello" }] },
+        redactStreamDiagnosticsForLog: true,
+        reqLogger: {
+          appendConvertedChunk() {
+            throw new Error(privateCue);
+          },
+        },
+      }
+    );
+  } finally {
+    console.log = originalLog;
+  }
+
+  const retainedDiagnostics = logCalls.flat().map(String).join(" ");
+  assert.equal(retainedDiagnostics.includes(privateCue), false);
+  assert.equal(retainedDiagnostics.includes("[omitted: video transcript]"), true);
+});

tests/unit/stream-readiness.test.ts

@@ -616,10 +616,7 @@ test("ensureStreamReadiness preserves sanitized error-only diagnostics on early
   assert.equal(result.response.status, 502);
   assert.equal(result.code, "STREAM_EARLY_EOF");
   assert.equal(result.type, "stream_early_eof");
-  assert.equal(
+  assert.equal(result.classificationReason, "Stream ended before producing a non-ping SSE event");
-    result.classificationReason,
-    "Stream ended before producing a non-ping SSE event"
-  );
   assert.equal(
     result.upstreamDiagnostic,
     "UPSTREAM_DETAIL quota exhausted; retry after 2s; empty content Bearer [REDACTED] <path>"
@@ -636,19 +633,38 @@ test("ensureStreamReadiness preserves sanitized error-only diagnostics on early
   assert.equal(body.upstream_details.error.message, result.upstreamDiagnostic);
   assert.equal(warnings.length, 1);
 
-  for (const surfaced of [
+  for (const surfaced of [result.reason, body.upstream_details.error.message, warnings[0]]) {
-    result.reason,
-    body.upstream_details.error.message,
-    warnings[0],
-  ]) {
     assert.match(surfaced, /UPSTREAM_DETAIL/);
-    assert.doesNotMatch(
+    assert.doesNotMatch(surfaced, /SECOND_DETAIL|TOP_SECRET|\/srv\/omniroute\/handler\.ts/);
-      surfaced,
-      /SECOND_DETAIL|TOP_SECRET|\/srv\/omniroute\/handler\.ts/
-    );
   }
 });
 
+test("ensureStreamReadiness omits sensitive upstream diagnostics from retained logs", async () => {
+  const rawCue = "PRIVATE_STREAM_READINESS_VIDEO_TRANSCRIPT_SENTINEL";
+  const warnings: string[] = [];
+  const response = new Response(
+    streamFromChunks([`data: ${JSON.stringify({ error: { message: rawCue } })}\n\n`]),
+    { status: 200, headers: { "Content-Type": "text/event-stream" } }
+  );
+
+  const result = await ensureStreamReadiness(response, {
+    timeoutMs: 100,
+    provider: "test-provider",
+    model: "test-model",
+    redactUpstreamDiagnosticForLog: true,
+    log: {
+      warn: (_tag, message) => warnings.push(message),
+    },
+  });
+
+  assert.equal(result.ok, false);
+  if (result.ok) assert.fail("error-only SSE payload must remain a readiness failure");
+  assert.match(result.reason, new RegExp(rawCue));
+  assert.equal(warnings.length, 1);
+  assert.doesNotMatch(warnings[0], new RegExp(rawCue));
+  assert.match(warnings[0], /upstream diagnostic omitted/);
+});
+
 test("stream-readiness diagnostics cannot reclassify Antigravity account exhaustion (#8972)", () => {
   const classificationError = "Stream ended before producing a non-ping SSE event";
   const diagnostic = "UPSTREAM_DETAIL quota exhausted; retry after 2s; empty content";

tests/unit/universal-handoff.test.ts

@@ -8,6 +8,7 @@ import {
   DEFAULT_UNIVERSAL_HANDOFF_CONFIG,
   type HandoffPayload,
 } from "../../open-sse/services/contextHandoff.ts";
+import { fingerprintVideoTranscriptDescription } from "../../src/lib/guardrails/videoTranscriptLogRedaction.ts";
 
 // ── resolveUniversalHandoffConfig ────────────────────────────────────────────
 
@@ -160,6 +161,49 @@ test("providerAllowlist: empty allowlist allows all providers", async () => {
   assert.ok(calls.length > 0, "handleSingleModel MUST be called when allowlist is empty");
 });
 
+test("transcript-sensitive model switches generate from a safely redacted history", async () => {
+  const sentinel = "PRIVATE_UNIVERSAL_HANDOFF_TRANSCRIPT_SENTINEL";
+  const description = `[Video description: stable scene; transcript[source=embedded;confidence=1.00;interval=00:01.000-00:02.000] text="${sentinel}"]`;
+  const calls: Array<{ body: Record<string, unknown>; modelStr: string }> = [];
+  maybeGenerateUniversalHandoff({
+    sessionId: "ses_private_video",
+    comboName: "private-video-combo",
+    messages: [
+      {
+        role: "user",
+        content: [
+          { type: "input_text", text: "Keep this universal handoff context" },
+          { type: "input_text", text: description },
+        ],
+      },
+    ],
+    prevModel: "openai/gpt-4o",
+    currModel: "anthropic/claude-3-5-sonnet",
+    videoTranscriptSensitive: true,
+    trustedDescriptionFingerprints: [fingerprintVideoTranscriptDescription(description)],
+    universalConfig: {
+      ...DEFAULT_UNIVERSAL_HANDOFF_CONFIG,
+      enabled: true,
+      providerAllowlist: [],
+      handoffModel: "anthropic/claude-3-5-sonnet",
+    },
+    handleSingleModel: async (body, modelStr) => {
+      calls.push({ body, modelStr });
+      return new Response(JSON.stringify({ choices: [{ message: { content: "{}" } }] }), {
+        status: 200,
+        headers: { "content-type": "application/json" },
+      });
+    },
+  });
+
+  await waitImmediate();
+  assert.strictEqual(calls.length, 1);
+  const retainedPrompt = JSON.stringify(calls[0].body);
+  assert.match(retainedPrompt, /Keep this universal handoff context/);
+  assert.match(retainedPrompt, /omitted: video transcript/);
+  assert.doesNotMatch(retainedPrompt, new RegExp(sentinel));
+});
+
 test("providerAllowlist: handoffModel takes precedence over currModel for allowlist check", async () => {
   const calls: unknown[] = [];
   await maybeGenerateUniversalHandoff({