debug

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of a process-oriented guide for developers and AI agents to systematically debug code issues. It defines a scientific workflow involving reproduction, hypothesis formation, and verification, which promotes safe and predictable behavior.
  • [COMMAND_EXECUTION]: The skill includes a standard command 'git diff HEAD~5' as a suggestion for gathering evidence by reviewing recent code changes. This is a common development operation and does not involve risky flags or unvalidated user input.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data such as error logs, console output, and reproduction steps.
  • Ingestion points: Instructions for gathering error logs and console output (SKILL.md).
  • Boundary markers: The skill suggests clear formatting for logging (e.g., 'Exact error message' blocks).
  • Capability inventory: Limited to shell execution for 'git diff' (SKILL.md).
  • Sanitization: None explicitly defined, but the skill's structure encourages treating error data as passive information for analysis rather than executable instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 11:19 PM
Security Audit — agent-trust-hub — debug