The Agent Skills Directory

[SAFE]: The skill implements a structured methodology for technical debugging through instructions and documentation designed to find root causes before implementing fixes. It follows standard development best practices.
[COMMAND_EXECUTION]: The skill provides a bash script find-polluter.sh that automates running npm test on local files to identify state-polluting tests. This is standard development utility behavior and occurs locally.
[DATA_EXFILTRATION]: Diagnostic examples in SKILL.md show how to verify environment variables and signing identities. These are documented techniques for local debugging of build processes and do not involve unauthorized data transmission or external exfiltration.
[PROMPT_INJECTION]: The skill guides the agent to process error logs and stack traces from the application environment. While this involves ingesting external data, it is a necessary part of the debugging process and is handled as intended functional logic.

systematic-debugging