9176571ec1
fix(gemini): sanitize thoughtSignatures for native Google provider
...
Native Google Gemini provider was accumulating 2K-8K tokens of Base64
thoughtSignature blobs per turn, causing premature context overflow.
The sanitizer was only enabled for OpenRouter Gemini, not native Google.
Fixes #23392
2026-02-22 12:24:53 +01:00
feccac6723
fix: sanitize thinking blocks for GitHub Copilot Claude models (openclaw#19459) thanks @jackheuberger
...
Verified:
- pnpm build
- pnpm check
- pnpm test:macmini
Co-authored-by: jackheuberger <12731288+jackheuberger@users.noreply.github.com>
Co-authored-by: Tak Hoffman <781889+Takhoffman@users.noreply.github.com>
2026-02-20 19:48:09 -06:00
b8b43175c5
style: align formatting with oxfmt 0.33
2026-02-18 01:34:35 +00:00
31f9be126c
style: run oxfmt and fix gate failures
2026-02-18 01:29:02 +00:00
d0cb8c19b2
chore: wtf.
2026-02-17 13:36:48 +09:00
ed11e93cf2
chore(format)
2026-02-16 23:20:16 -05:00
90ef2d6bdf
chore: Update formatting.
2026-02-17 09:18:40 +09:00
68ea063958
🤖 fix: preserve openai reasoning replay ids ( #17792 )
...
What:
- disable tool-call id sanitization for OpenAI/OpenAI Codex transcript policy
- gate id sanitization in image sanitizer to full mode only
- keep orphan reasoning downgrade scoped to OpenAI model-switch replay path
- update transcript policy, session-history, sanitizer, and reasoning replay tests
- document OpenAI model-switch orphan-reasoning cleanup behavior in transcript hygiene reference
Why:
- OpenAI Responses replay depends on canonical call_id|fc_id pairings for reasoning followers
- strict id rewriting in OpenAI path breaks follower matching and triggers rs_* orphan 400s
- limiting scope avoids behavior expansion while fixing the identified regression
Tests:
- pnpm vitest run src/agents/transcript-policy.test.ts src/agents/pi-embedded-runner.sanitize-session-history.test.ts src/agents/openai-responses.reasoning-replay.test.ts
- pnpm vitest run --config vitest.e2e.config.ts src/agents/transcript-policy.e2e.test.ts src/agents/pi-embedded-runner.sanitize-session-history.e2e.test.ts src/agents/pi-embedded-helpers.sanitize-session-messages-images.removes-empty-assistant-text-blocks-but-preserves.e2e.test.ts src/agents/pi-embedded-helpers.sanitizeuserfacingtext.e2e.test.ts
- pnpm lint
- pnpm format:check
- pnpm check:docs
- pnpm test (fails in current macOS bash 3.2 env at test/git-hooks-pre-commit.integration.test.ts: mapfile not found)
2026-02-15 22:45:01 -08:00
94763cd87d
Fix OpenAI/Codex tool call id sanitization for transcript policy ( #15279 )
2026-02-13 11:39:51 +00:00
1d2c5783fd
fix(agents): enable tool call ID sanitization for Anthropic provider ( #13830 )
...
Co-authored-by: 0xRaini <0xRaini@users.noreply.github.com>
2026-02-12 08:42:24 +09:00
f06dd8df06
chore: Enable "experimentalSortImports" in Oxfmt and reformat all imorts.
2026-02-01 10:03:47 +09:00
5ceff756e1
chore: Enable "curly" rule to avoid single-statement if confusion/errors.
2026-01-31 16:19:20 +09:00
2496056886
fix(minimax): use correct API endpoint and format
...
MiniMax has updated their API. The previous configuration used an
incorrect endpoint (api.minimax.io/anthropic) with anthropic-messages
format, which no longer works.
Changes:
- Update MINIMAX_API_BASE_URL to https://api.minimax.chat/v1
- Change API format from anthropic-messages to openai-completions
- Remove minimax from isAnthropicApi check in transcript-policy
This fixes the issue where MiniMax API calls return no results.
2026-01-27 19:48:38 -06:00
2424404fb4
docs: add transcript hygiene reference
2026-01-23 01:34:21 +00:00
db0235a26a
fix: gate transcript sanitization by provider
2026-01-23 00:42:45 +00:00
echoVic
jackheuberger
Peter Steinberger
cpojer
Sebastian
the sun gif man
JINNYEONG KIM
0xRain
hlbbbbbbb