The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from GitHub pull requests, issues, and repository documentation to automate its review and merge tasks.\n- Ingestion points: The agent is instructed in SKILL.md to read repository instructions, issue content, acceptance criteria, PR metadata, and code diffs.\n- Boundary markers: There are no explicit instructions, delimiters, or safety warnings provided to the agent to help it distinguish between legitimate data and potential instructions embedded within the content it reviews.\n- Capability inventory: The skill uses the scripts/approve_or_merge_pr.sh script to perform high-privilege operations, such as approving and merging pull requests, based on its findings.\n- Sanitization: No sanitization, filtering, or validation of the retrieved external data is performed before the agent processes it or takes action.

user-story-reviewer