The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute standard development commands such as npm run build, npm run test, npx tsc, and git diff to assess project health.
[SAFE]: The skill includes a proactive security phase that uses grep locally to identify potential hardcoded secrets (e.g., API keys with the sk- prefix) and unintended debug logs in the source code.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes external data (tool outputs, test results, and file diffs) which could contain adversarial instructions intended to misguide the agent's summary.

verification-loop