The Agent Skills Directory

[SAFE]: The skill implements a systematic debugging framework based on established industry practices (Kernighan & Pike, Hunt & Thomas). This methodology prioritizes rigorous verification and isolation, which inherently reduces the risk of introducing accidental vulnerabilities during the debugging process.
[COMMAND_EXECUTION]: The skill references standard development tools such as git bisect, xxd, and head for diagnostic purposes. These are used in a limited, controlled manner to identify bug loci (e.g., hex-dumping file headers to detect UTF-8 BOM issues).
[SAFE]: It provides a specific Python one-liner for hex-dumping file content: python3 -c "print(open('file.csv', 'rb').read(8).hex())". This is a fixed diagnostic command used to inspect data rather than execute untrusted input.
[EXTERNAL_DOWNLOADS]: The skill includes citations to reputable developer resources (e.g., alejandromp.com, sarunw.com) to provide context for specific debugging scenarios (SwiftUI layout issues). These links are educational and do not target executable scripts or untrusted repositories.
[SAFE]: The skill explicitly warns against security-degrading anti-patterns such as masking exceptions with broad try/except blocks or ignoring silent data corruption, instead directing the agent to root-cause tracing and defense-in-depth layering.

systematic-debugging