shumengya/sproutclaw
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs(changelog): audit unreleased entries

Browse Source
This commit is contained in:
Mario Zechner
2026-03-20 02:04:22 +01:00
parent e3fee7a511
commit a2baeb9a9d
2 changed files with 31 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
packages/ai/CHANGELOG.md
View File

@@ -4,12 +4,14 @@
### Added
- Added `gpt-5.4-mini` model support for the `openai-codex` provider with Codex pricing metadata and unit coverage.
- Added `gpt-5.4-mini` model support for the `openai-codex` provider with Codex pricing metadata and unit coverage ([#2334](https://github.com/badlogic/pi-mono/pull/2334) by [@justram](https://github.com/justram))
### Fixed
- Fixed `validateToolArguments()` to fall back gracefully when AJV schema compilation is blocked in restricted runtimes such as Cloudflare Workers, allowing tool execution to proceed without schema validation ([#2395](https://github.com/badlogic/pi-mono/issues/2395))
- Fixed `google-vertex` API key resolution to ignore placeholder auth markers like `<authenticated>` and fall back to ADC instead of sending them as literal API keys ([#2335](https://github.com/badlogic/pi-mono/issues/2335))
- Fixed OpenRouter reasoning requests to use the provider's nested `reasoning.effort` payload instead of OpenAI's `reasoning_effort`, restoring thinking level support for OpenRouter models ([#2298](https://github.com/badlogic/pi-mono/pull/2298) by [@PriNova](https://github.com/PriNova))
- Fixed Bedrock prompt caching for application inference profiles by allowing cache points to be forced with `AWS_BEDROCK_FORCE_CACHE=1` when the profile ARN does not expose the underlying Claude model name ([#2346](https://github.com/badlogic/pi-mono/pull/2346) by [@haoqixu](https://github.com/haoqixu))
## [0.60.0] - 2026-03-18