Skip to content

fix: pin @pierre/diffs to exact version 1.1.12#614

Open
xiaolai wants to merge 1 commit intobacknotprop:mainfrom
xiaolai:fix/nlpm-pin-pierre-diffs-version
Open

fix: pin @pierre/diffs to exact version 1.1.12#614
xiaolai wants to merge 1 commit intobacknotprop:mainfrom
xiaolai:fix/nlpm-pin-pierre-diffs-version

Conversation

@xiaolai
Copy link
Copy Markdown

@xiaolai xiaolai commented Apr 26, 2026

Automated audit: This PR was generated by NLPM, a natural language programming linter, running via claude-code-action. Please evaluate the diff on its merits.

Issue

@pierre/diffs is currently declared with a caret range (^1.1.12) in package.json, which permits automatic minor and patch version upgrades without explicit review.

This package is particularly sensitive: the pierre-guard skill in this repo explicitly documents that @pierre/diffs uses shadow DOM selectors and grid layout assumptions that can break silently on minor version bumps. An unreviewed upgrade could cause subtle rendering regressions that bypass automated testing.

Fix

Pin @pierre/diffs to the exact version 1.1.12. Future upgrades should be run through the pierre-guard verification checklist before the pin is advanced.

The other three SDK dependencies (@anthropic-ai/claude-agent-sdk, @openai/codex-sdk, @opencode-ai/sdk) are externally versioned and benefit from caret ranges for security patches; this change deliberately targets only @pierre/diffs which carries unique internal breakage risk per the guard skill.

The pierre-guard skill explicitly documents that @pierre/diffs uses shadow
DOM selectors and grid assumptions that can break silently on minor upgrades.
Pinning to an exact version ensures upgrades are intentional and gated
through the pierre-guard verification checklist.

Co-Authored-By: Claude Code <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants