The Agent Skills Directory

[SAFE]: The skill instructions describe standard, disciplined software debugging and performance analysis workflows.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process bug reports, error logs, and captured network traces. While these external data sources could theoretically contain prompt injection attempts, the skill's structured approach (reproduce -> hypothesize -> verify) provides a natural layer of defense against accidental obedience to instructions found within debug data.
[DYNAMIC_EXECUTION]: The skill involves creating and running reproduction scripts and tests. This is the primary intended purpose of the skill and is carried out using standard development tools (Bash, Playwright, test runners) within the agent's environment.

diagnose