The Agent Skills Directory

[SAFE]: The skill provides a structured methodology for debugging code failures. It focuses on systematic investigation, hypothesis testing, and architectural review.\n- [NO_CODE]: The skill consists primarily of natural language instructions and shell-script instrumentation examples for diagnostic purposes. It does not include any executable binary files or complex scripts that could pose a security risk.\n- [COMMAND_EXECUTION]: The skill includes benign examples of shell commands used for logging environment variables and state (e.g., echo, env, grep). These are documented as instrumentation patterns for debugging and do not incorporate unvalidated user input or remote code.\n- [REMOTE_CODE_EXECUTION]: No remote code execution patterns were detected. The skill references an upstream GitHub repository for metadata and sourcing information but does not download or execute remote scripts.\n- [EXTERNAL_DOWNLOADS]: The skill refers to its source on GitHub (obra/superpowers), which is a public documentation/skill repository. No automated downloads or package installations are triggered by this skill.

systematic-debugging