Files
OmniRoute/open-sse/handlers/chatCore/targetFormat.ts
GiauPhan 548316a2c4 fix(translator): Normalize tool call names from lowercase to PascalCase when translating upstream responses to Claude Messages API format (#10392)
* fix(translator): Normalize tool call names from lowercase to PascalCase (#1)

* Fix: Map lowercase tool names from Antigravity (Gemini format) to Claude Code expected PascalCase

* Fix: toolNameMap in fun restoreClaudePassthroughToolUseName

* fix(translator): Normalize tool call names from lowercase to PascalCase when translating upstream responses (OpenAI, Gemini, Antigravity) to Claude Messages API format

This resolves `Error: No such tool available: read`/`bash`/`write` errors when using Claude Code CLI with third-party providers that emit lowercase tool names. The fix adds case-insensitive tool name lookups in `openai-to-claude.ts`, `gemini-to-claude.ts`, and related translators, ensuring tool names like `read`/`bash` are mapped to `Read`/`Bash` before being sent to Claude Code. Includes unit tests and comprehensive changelog notes ([#10250](https://github.com/diegosouzapw/OmniRoute/pull/10250))

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

* fix(translator): Parse <tool_call> JSON and TOOL_CALL text formats fr… (#2)

* fix(translator): Parse <tool_call> JSON and TOOL_CALL text formats from model output

Some models (DeepSeek, Qwen) emit tool calls as text instead of proper
tool_calls JSON: either <tool_call>{...}</tool_call> or TOOL_CALL Name: {...}.
Extend extractXmlInvokeBlocks to handle all 3 formats in a single scan pass,
picking whichever pattern appears first. Includes unit tests for all formats.

* fix(translator): Parse text-format tool calls in gemini-to-claude translator

Extend the Gemini->Claude translator to detect <invoke>, <tool_call> JSON,
and TOOL_CALL text formats emitted inline in text parts (Antigravity/Gemini
models), converting them to proper tool_use content blocks instead of leaking
raw text to Claude Code.

* docs(changelog): Add changelog entry for text tool call parsing fix

* fix(translator): consolidate tool name casing normalization and restore thought-signature persistence (#3)

* fix(translator): sanitize tool_use.id and tool_result.tool_use_id to match Anthropic schema (#4)

Ensure tool IDs from OpenAI-compatible upstreams (which may contain dots, colons, or special characters) are sanitized to ^[a-zA-Z0-9_-]+$ in response translators and passthrough requests before reaching Claude endpoints.

* fix(responses): preserve native tools for openai-compatible Responses targets (#5)

A Responses-shaped request to a custom openai-compatible connection whose
outbound protocol is Responses took a Responses -> Chat -> Responses round
trip, so Codex custom tools lost their grammar (`exec`), namespace groups were
flattened (`collaboration`), and tool invocations failed upstream.

Gate a native Responses passthrough on the connection's configured protocol
(`apiType: "responses"` / `_omnirouteForceResponsesUpstream`) so the original
tool definitions reach a Responses-capable upstream unchanged. Chat-only
connections keep the existing downgrade.

Closes #10374

* fix(translator): add support for 'applypatch' tool name in tool call checks

* test(translator): add unit test for apply_patch and applypatch tool name remapping

* fix(translator): remove no-explicit-any lint errors in tool-use-id-sanitization test

Type the openaiToClaudeResponse/translateNonStreamingResponse return
values with narrow local shapes instead of `any`, satisfying the
repo's no-explicit-any = error rule for tests/. No behavior change —
the same 3 assertions still pass.

Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>

* test: update 9568 casing regression to match #10392's consolidated fix

restoreClaudeToolName's static casing map now normalizes known
lowercase tool names to canonical PascalCase unconditionally on the
gemini-to-claude and openai-to-claude Claude Messages API paths (not
gated behind toolNameMap), superseding the earlier per-map-only fix
that the original #9568 regression test locked in as "expected" (it
was previously labeled a known bug case). The gemini-to-openai
passthrough path is unaffected by #10392 and keeps its original
pass-through assertion.

Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>

---------

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
2026-08-18 10:49:48 -03:00

80 lines
3.6 KiB
TypeScript

/**
* chatCore wire target-format resolver (Quality Gate v2 / Fase 9 — chatCore god-file
* decomposition, #3501).
*
* Pure resolution of the provider alias + the upstream target format used to translate the request.
* Model/custom overrides win first. A declared connection-level alternate protocol wins next. A
* Responses-shaped inbound request otherwise keeps the Responses wire format, except for custom
* OpenAI-compatible connections explicitly configured for Chat.
* AgentRouter may inherit the inbound protocol when no explicit connection override exists.
* Returns both `alias` (reused by the handler when stripping the `alias/` prefix off the upstream
* model id) and `targetFormat`.
*/
import { PROVIDER_ID_TO_ALIAS, getModelTargetFormat } from "../../config/providerModels.ts";
import { getRegistryEntry } from "../../config/providerRegistry.ts";
import { resolveAlternateFormat } from "../../config/providers/alternateFormats.ts";
import { getTargetFormat } from "../../services/provider.ts";
import { FORMATS } from "../../translator/formats.ts";
export function resolveChatCoreTargetFormat(opts: {
provider: string;
resolvedModel: string;
apiFormat: string | undefined;
sourceFormat?: string;
customModelTargetFormat: string | undefined;
providerSpecificData: unknown;
nativeXaiResponsesPassthrough?: boolean;
nativeOpenAICompatibleResponsesPassthrough?: boolean;
}) {
const {
provider,
resolvedModel,
apiFormat,
sourceFormat,
customModelTargetFormat,
providerSpecificData,
nativeXaiResponsesPassthrough = false,
nativeOpenAICompatibleResponsesPassthrough = false,
} = opts;
const alias = PROVIDER_ID_TO_ALIAS[provider] || provider;
const modelTargetFormat = getModelTargetFormat(alias, resolvedModel);
const explicitConnectionTargetFormat = (
providerSpecificData as { targetFormat?: unknown } | null | undefined
)?.targetFormat;
const inferredAgentRouterTargetFormat =
provider === "agentrouter" &&
!(typeof explicitConnectionTargetFormat === "string" && explicitConnectionTargetFormat) &&
(sourceFormat === FORMATS.OPENAI_RESPONSES ||
sourceFormat === FORMATS.OPENAI ||
sourceFormat === FORMATS.CLAUDE)
? sourceFormat
: undefined;
const providerTargetFormat = getTargetFormat(provider, providerSpecificData);
const declaredConnectionAlternate = resolveAlternateFormat(
getRegistryEntry(provider),
providerSpecificData
);
const customOpenAICompatible = provider.startsWith("openai-compatible-");
// #8994: model-level targetFormat overrides (from registry or custom-model DB override)
// take precedence over apiFormat="responses" — otherwise Vertex Claude models with
// targetFormat="claude" get wrongly routed to OpenAI Responses format.
// #9161: a custom OpenAI-compatible Chat connection must likewise keep its configured
// outbound protocol when a Responses-shaped client (for example Codex) calls /responses.
// Registry-declared connection alternates are equally explicit: a DeepSeek connection set to
// Anthropic must stay on /anthropic/v1/messages even when the caller speaks Responses.
let targetFormat =
modelTargetFormat ||
customModelTargetFormat ||
declaredConnectionAlternate?.format ||
(apiFormat === "responses" && !customOpenAICompatible
? FORMATS.OPENAI_RESPONSES
: inferredAgentRouterTargetFormat || providerTargetFormat);
if (nativeXaiResponsesPassthrough || nativeOpenAICompatibleResponsesPassthrough) {
targetFormat = FORMATS.OPENAI_RESPONSES;
}
return { alias, targetFormat };
}
export type ChatCoreTargetFormat = ReturnType<typeof resolveChatCoreTargetFormat>;