pr-review
Review someone else's GitHub PR - checkout locally, deep read-only review, output copyable findings. Never posts, approves, or merges. Invoke when Josh says things like 'review this PR', 'can you look at Dave's PR', 'take a look at #12', 'review PR 340 for me', 'what do you think of this pull request', 'look over their changes', or drops a PR link and asks for a look. Disambiguation with pr-feedback: authorship decides. A PR authored by someone else belongs here. A PR Josh authored that has review comments to act on belongs to pr-feedback - cues for that are 'our PR', 'my PR', 'feedback we got', 'address the review', 'reviewer said'. When the phrasing is ambiguous ('look at PR 12'), check the PR author with gh before choosing. Takes a PR ref plus an optional note; with no ref, resolves the PR from the conversation or the current branch.
90%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
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 required workflow reads GitHub PRs, comments, and linked issues authored by third parties.
- 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.