pr-feedback
Address review feedback on a PR Josh authored - fetch the review comments, understand them, implement the fixes locally, and resolve the threads it addressed once Josh has reviewed, approved, and pushed. Never posts on his behalf. Invoke when Josh says things like 'we got some feedback on our PR', 'address the review comments', 'the reviewer said X', 'my PR has comments on it', 'can you handle the feedback on #12', 'fix what the review flagged', or mentions someone reviewing a PR of his. Disambiguation with pr-review: authorship decides. A PR Josh authored with review comments to act on belongs here - cues are 'our PR', 'my PR', 'feedback we got', 'address the review', 'reviewer said'. A PR authored by someone else that he wants read belongs to pr-review - cues are 'review this PR', "take a look at Dave's PR", 'can you review #12'. When the phrasing is ambiguous ('look at PR 12'), check the PR author with gh before choosing. Takes an optional PR ref and note; with no ref, resolves the PR from the conversation or the current branch.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
The skill reads pull request review comments, issue comments, and review threads via GitHub CLI (`gh api` and GraphQL) in Step 4, which ingests outsider-authored free text from PR reviewers.
- Repository
joshmanders/dotfiles
- Revision
92186fe
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.