The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use various shell commands and repository-specific scripts to manage Pull Requests.
Evidence: Directs the usage of git for rebasing and pushing, make for local testing, and gh for commenting on PRs.
Evidence: Invokes multiple repository-specific scripts in the ./scripts/ directory for review management and status checking.
Evidence: Instructs the agent to access environment variables via bash to populate attribution footers.
[INDIRECT_PROMPT_INJECTION]: The skill processes instructions from an external automated reviewer, creating a potential surface for indirect injection.
Ingestion points: Reads review comments from the Codex system using ./scripts/check_codex_comments.sh.
Boundary markers: No specific delimiters or "ignore warnings" are defined for the comment content.
Capability inventory: The agent has capabilities to push code changes (git push) and post comments (gh pr comment).
Sanitization: There are no explicit instructions for sanitizing or filtering the external comment content.

pull-requests