Published to npm as @bitkyc08/opencodex@2.22.0 with dist-tag latest.
New Features
- Add complete French localization (#1733)
- Inspect Codex diagnostic logs safely - Part 1 (#1727)
- Protect Codex diagnostic logs - Part 2 (#1729)
- Reclaim - Part 3 (#1732)
- Keep ChatGPT-native models on v1 when catalog is v2 (#1728)
- Probe Command Code 5h/weekly credits (#1761)
Bug Fixes
- Default store:false on /v1/responses inbound when omitted (#1743)
- Source capability evidence from explicit catalog provenance (#1799)
- Harden quality and duplicate closure (#1764)
- Fix release changelog coverage across previews and direct commits (#1740)
- Allow slash custom model IDs and live Cloudflare Workers AI search (#1793)
- Close validated CodeQL parser and quality-gate findings (#1750)
- Ingest llama.cpp multimodal and dual-envelope metadata (#1807)
Other Changes
- Merge dev into main: v2.22.0 release (b0e0ff8f)
Changelog
Full Changelog: v2.21.0...v2.22.0
- #1727 feat(storage): inspect Codex diagnostic logs safely - Part 1 @Wibias
- #1728 feat(agents): keep ChatGPT-native models on v1 when catalog is v2 @jonathanli12
- #1729 feat(storage): protect Codex diagnostic logs - Part 2 @Wibias
- #1732 feat(log-guard): Reclaim - Part 3 @Wibias
- #1733 feat(i18n): add complete French localization @ulrich-zogo
- #1740 Fix release changelog coverage across previews and direct commits @Wibias
- #1743 fix(server): default store:false on /v1/responses inbound when omitted @SOSANA
- #1750 fix: close validated CodeQL parser and quality-gate findings @Wibias
- #1761 feat(quota): probe Command Code 5h/weekly credits @yansigit
- #1764 fix(issue-triage): harden quality and duplicate closure @Wibias
- #1793 fix: allow slash custom model IDs and live Cloudflare Workers AI search @yansigit
- #1799 fix(routing): source capability evidence from explicit catalog provenance @lidge-jun
- #1807 fix(catalog): ingest llama.cpp multimodal and dual-envelope metadata @lidge-jun
- b0e0ff8f Merge dev into main: v2.22.0 release