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.
The required workflow runs `await_review.py`, which returns the PR review body and state from GitHub, and SKILL.md then directs the agent to read the review and inline threads, which are PR text that outside users can post on public repos without the agent selecting a specific item.
jbaruch/skill-issue-policy
60ffdd8
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.