The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard shell commands to verify the build status and code quality of the project.
Evidence: npm run build, pnpm build, npx tsc, pyright ., ruff check ., and npm run test are used to run development toolchains.
[CREDENTIALS_UNSAFE]: The skill performs a local security scan to identify hardcoded secrets in the source code.
Evidence: Uses grep to search for patterns like sk- and api_key in JavaScript and TypeScript files. This is a defensive security measure.
[COMMAND_EXECUTION]: The skill utilizes Git commands to review changes before a pull request.
Evidence: git diff --stat and git diff HEAD~1 are used to provide a summary of modified files.
[PROMPT_INJECTION]: The skill processes output from build logs and test results, which represents a potential surface for indirect prompt injection if those outputs contain malicious instructions. However, the capabilities are restricted to standard development workflows.

verification-loop