- Normalize tool schemas: strip additionalProperties and empty required arrays
- Merge consecutive assistant messages and adjacent user turns after role normalization
- Prepend synthetic user message when conversation starts with assistant
- Convert orphaned toolResults to inline text when assistant with toolUses is missing
- Enforce strictly alternating user/assistant roles in history
- Use deterministic uuidv5 conversationId based on first message for session caching
- Ensure origin field is present on all userInputMessage entries
- Resolve <<<<<<< HEAD conflict in RoutingTab.tsx by keeping the
HEAD version with aria-disabled and pre-computed titleText
- Fix inconsistent indentation in CLI help text (providers commands
and CLI Tools section)
After cherry-picking PR #2231, the function signature changed from
positional (provider, model) to object ({ provider, model }). Fixes the
2 pre-existing tests that still used the old positional style.
- chatCore.ts: pass {requestId:skillRequestId,messageIndex:0} to cacheReasoningFromAssistantMessage
- responseSanitizer.ts: widen isDeepSeekV4Model regex to match all deepseek-v4 variants
Fixes Anthropic HTTP 400 errors (~49/h on claude-opus-4-7) by preserving
the latest assistant message's thinking blocks verbatim instead of
rewriting them to redacted_thinking.
Co-authored-by: NomenAK <anton@nomenak.dev>
The root package.json was updated in 52f3285d to drop Node 20.x support
(http-proxy-middleware 4.x requirement). electron/package.json had no
engines field declared, leaving the desktop build implicitly permissive.
Adds the same constraint (>=22.22.2 <23 || >=24.0.0 <27) to keep the
electron workspace consistent with the root engine policy.
Refs: #2228
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
http-proxy-middleware 4.x (introduced via #2228) requires Node >=22.15.0.
Updated engines.node to >=22.22.2 <23 || >=24.0.0 <27 (drops 20.x).
BREAKING CHANGE: users on Node 20.x must upgrade to Node 22.22.2+ or 24+.
Refs: #2228
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- antigravity: AntigravityCredentials.projectId widened to string|null
to match base ProviderCredentials shape post-#2227 squash merge.
- responses-handler: heartbeat assertion updated for #2233's new
openai-responses-in-progress shape (was: keepalive comment).
- search-registry: expected count is now 12 (ollama-search +
zai-search both landed in this release).
After merging PRs #2221 (ModelSync shared loopback readiness gate + IPv4 force)
and #2219 (Antigravity loadCodeAssist bootstrap + fetchAvailableModels fallback)
into release/v3.8.0, two test suites needed updates to match the new routing:
- tests/unit/model-sync-route.test.ts:
* resetStorage() now calls __resetLoopbackReadinessForTests() so the
module-level __loopbackReadyPromise cache does not leak between tests.
* Every fetch mock now answers the /__readiness_probe__/ URL with 404 so
the gate opens immediately (any HTTP response satisfies the probe).
* Self-fetch target URL assertions updated from http://localhost/...
to http://127.0.0.1:20128/... per PR #2221's IPv4-force.
- tests/unit/provider-models-route.test.ts:
* The Antigravity discovery-retry test now treats loadCodeAssist calls as
non-fatal failures so the discovery path is still exercised.
* The expected discovery URL sequence is updated to the new
fetchAvailableModels-first order introduced by PR #2219.
The local-aliases-precedence path used `typeof aliases[parsed.model] === "string"`
to guard string-only operations, but TypeScript does not narrow the variable
`directTarget` from that index-expression test — the variable retained the union
type ModelAliasValue (string | object), so `indexOf`/`slice` were typed as
property accesses on the object branch and the strict-core typecheck failed.
Refactors to capture `directTarget` first and run `typeof directTarget === "string"`
on the variable, which TS does narrow. No runtime semantics change — local-aliases
tests still pass.
Wires the new platform-overhaul gates into the Detailed Checklist used by
the release-cut workflow:
Documentation block:
npm run check:docs-all (umbrella over sync, counts, env-doc, deprecated, doc-links)
npm run check:env-doc-sync (code ↔ .env.example ↔ ENVIRONMENT.md parity)
npm run check:doc-links (no broken internal markdown refs)
i18n block (replaces stale `scripts/i18n-check.mjs` mention):
npm run i18n:check (drift between source docs and .i18n-state.json)
npm run i18n:check-ui-coverage (every locale ≥ 80% UI key coverage)
npm run i18n:sync-ui:dry (0 missing keys across 40 locales)
Note about running npm run i18n:run when source English docs change.
These are the same checks newly enforced by the docs-sync-strict and
i18n-ui-coverage CI jobs added in commit acf6b93d.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
