The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from external GitHub issues which could contain instructions intended to manipulate the agent's behavior.
Ingestion points: GitHub issue titles and descriptions fetched via gh issue view in SKILL.md.
Boundary markers: None; the external content is integrated into the agent's context without clear delimiters or instructions to disregard embedded commands.
Capability inventory: The skill has broad capabilities including file modification (applying fixes), shell command execution (pnpm, git), and PR creation (gh pr create).
Sanitization: No validation or sanitization is performed on the data retrieved from GitHub before it is processed.
[COMMAND_EXECUTION]: The skill makes extensive use of system commands to perform development tasks.
Evidence: Executes git checkout, git pull, pnpm test, pnpm lint, gh pr create, and gh run watch in SKILL.md to manage the codebase and CI/CD status.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to sync with remote repositories and services.
Evidence: Uses git fetch, git push, and various gh (GitHub CLI) commands to interact with the GitHub platform. These interactions are directed at a well-known service (GitHub).

hotfix