The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions utilize standard development tools such as pnpm, ng, and curl to establish feedback loops for debugging. These are executed locally within the user's project environment.
[SAFE]: The skill includes explicit instructions for security best practices, such as redacting secrets and credentials from logs and terminal output before displaying them to the user.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided bug reports and symptoms to guide the debugging process. While it involves executing commands based on the context of the bug, the workflow emphasizes a structured, step-by-step approach that requires building a deterministic feedback loop before proceeding, which reduces the risk of accidental execution of malicious input.

diagnosing-bugs