systematic-debugging
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of documentation and reference guides for debugging workflows. No malicious code, obfuscation, or unauthorized access patterns were detected.\n- [COMMAND_EXECUTION]: The reference file 'references/debugging-techniques.md' contains example shell commands for debugging, such as 'git bisect', 'diff', 'env', and 'docker exec'. These are provided as educational content for environment comparison and diagnostic tracing and do not involve automated execution.\n- [PROMPT_INJECTION]: The debugging process involves analyzing external data (e.g., error logs, stack traces), which constitutes a potential surface for indirect prompt injection. However, this is inherent to the debugging task and not a flaw in the skill.\n
- Ingestion points: System logs, error messages, and stack traces described in 'SKILL.md'.\n
- Boundary markers: Not present.\n
- Capability inventory: Mentions of diagnostic shell commands ('git', 'env', 'diff') and file reads/writes for logging.\n
- Sanitization: No instructions are included to sanitize or escape log content before analysis.
Audit Metadata