The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using git, gh (GitHub CLI), and glab (GitLab CLI) to perform repository operations, fetch PR metadata, and submit replies.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and acts upon untrusted external data. 1. Ingestion points: The skill reads pull request comments and review bodies via the gh api and glab mr view commands. 2. Boundary markers: There are no defined delimiters or instructions provided to the agent to treat PR comment text as data rather than instructions. 3. Capability inventory: The agent has the ability to modify source code, execute git push (potentially with --force-with-lease), and interact with forge APIs to resolve threads and add reviewers. 4. Sanitization: No sanitization or filtering logic is present to handle potentially malicious instructions embedded within PR comments or reviews.

pr-ready