The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands including grep, find, and gh to identify testing, linting, and CI/CD configurations.
[REMOTE_CODE_EXECUTION]: The skill attempts to run pnpm coverage or npm run coverage, which executes scripts defined in the repository's package.json. This is expected behavior for a quality audit tool interacting with a local codebase.
[PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface by processing data from repository files and API responses.
Ingestion points: Reads package.json, tsconfig.json, and output from the gh api command.
Boundary markers: Absent.
Capability inventory: Includes shell command execution and invocation of local scripts.
Sanitization: No validation or sanitization is applied to the ingested data before it is used to generate findings.

check-quality