The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use standard CLI tools including npx playwright, gh, and ls to gather diagnostic information such as test reports, traces, and CI logs. These commands are typical for automated testing workflows and are well-scoped to the skill's primary purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from test artifacts, console logs, and CI metadata. It mitigates potential injection risks by instructing the agent to quote raw values directly, cite concrete evidence locations, and avoid paraphrasing content. Additionally, it maintains a strict human-in-the-loop requirement for the execution of any proposed code modifications or test quarantines.

Flaky Test Doctor