Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coherent stacked-PR workflow helper, with disclosed PR-reading and root-summary commenting behavior but broad triggers users should be aware of.
Install this if you want stack-aware PR review or fix workflows. Before using it, verify that your GitHub CLI account has only the repository permissions you intend, and be aware that stack mode can post or update a consolidated comment on the root PR. The broad pr and review triggers may be noisier than ideal, so use explicit --stack workflows when possible.
Using 'review' as a trigger can shadow or collide with a built-in review command, causing this skill to activate in contexts where a user intended a standard review workflow. In this skill's context, unintended activation matters because it changes behavior from single-PR handling to stack detection/iteration and can lead to unexpected PR comment posting across multiple PRs.
Overly Broad Trigger: 'pr' is too short and may match unintended inputs
The manifest description frames the skill as detecting shared stack membership and iterating a command across PRs. However, the documented contract also requires writing a consolidated comment to the root PR via gh pr comment, which is a distinct side effect not reflected in that description.
The note says the skill is a 'read/orchestration skill' and emphasizes that it does not push, rebase, or edit branches. Earlier instructions nevertheless require posting a summary comment on the root PR, so the documentation understates the skill's write effects and may mislead readers into expecting a read-only orchestration role.
No suspicious patterns detected.
