The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several shell commands to manage the development workflow, including gh for GitHub API interactions, git for repository management, poetry and pnpm for running backend/frontend tests and formatting tools, and curl for checking local service health.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external GitHub contributors which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Pull request descriptions (gh pr view) and various types of comments (inline threads, top-level reviews, and conversation comments) fetched via the GitHub API.
Boundary markers: The skill does not employ delimiters or specific instructions to help the agent distinguish between its own system instructions and the content of the external comments it processes.
Capability inventory: The agent has the authority to write and commit code, push changes to the repository, execute project-specific scripts via build tools, and perform authenticated network operations through the GitHub CLI.
Sanitization: There is no evidence of sanitization, validation, or escaping of the comment content before it is used as context for the agent to generate code fixes.

pr-address