The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from GitHub pull request comments, issue comments, and reviews, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Data retrieved via gh api from pulls/{n}/comments, issues/{n}/comments, and pulls/{n}/reviews in SKILL.md.
Boundary markers: No explicit delimiters or boundary markers are defined for the ingested data.
Capability inventory: The agent is instructed to fix bugs, correctness issues, and security vulnerabilities, which involves code modification and pushing changes to the repository.
Sanitization: No explicit sanitization or escaping of the comment content is mentioned, although the skill instructs the agent to "Judge each comment on its merits" and "Check them against the surrounding code."
[COMMAND_EXECUTION]: The skill utilizes the GitHub CLI tool (gh api) to fetch pull request and issue metadata. This is a standard and expected operation for tasks involving GitHub integration.

review-bots