The Agent Skills Directory

[SAFE]: The skill is purely instructional and provides a structured framework for debugging (Stop-the-Line, Triage Checklist) without containing executable malicious code.
[INDIRECT_PROMPT_INJECTION]: The skill includes defensive instructions specifically addressing indirect prompt injection risks by advising the agent to treat error output as untrusted data and not follow instructions or URLs embedded in stack traces or logs.
[COMMAND_EXECUTION]: The skill references standard development tools such as git bisect and npm test for diagnostic purposes within the developer's local environment, which is consistent with its stated purpose of debugging.
[NO_OBFUSCATION]: No obfuscated content, hidden characters, or deceptive metadata were detected in the skill instructions.

debugging-and-error-recovery