The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests test annotations from the codebase to generate reports. While this is an attack surface, the processing is restricted to specific date-based patterns, and the agent's logic is primarily focused on identifying expired dates rather than executing content.
[COMMAND_EXECUTION]: The skill provides a Bash script and a GitHub Actions workflow for CI integration, using standard tools like grep, awk, and the GitHub CLI for local repository management and issue creation.
[SAFE]: All external URLs and resources belong to trusted organizations or well-known services (e.g., Google Testing Blog, Playwright, and official GitHub Actions). All operations are local to the user's environment or CI pipeline.

flaky-test-quarantine