The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from GitHub PR comments.
Ingestion points: The gh api call in SKILL.md fetches comment bodies which are external, user-generated content.
Boundary markers: There are no explicit markers or instructions to treat the comment text as data rather than instructions.
Capability inventory: The agent has the ability to modify source code and execute GitHub API calls (replies/resolutions) based on these comments.
Sanitization: No sanitization or filtering is performed on the incoming comment text before analysis.
[COMMAND_EXECUTION]: The skill uses the GitHub CLI tool to perform its operations.
Evidence: Uses gh pr view, gh api, and gh api graphql throughout SKILL.md to interact with the repository and pull requests.

resolve-comments