debug-agent

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary purpose is bug diagnosis and fixing, and its instructions are strictly confined to these tasks without attempting to bypass safety constraints or access unauthorized data.
  • [SAFE]: The resource files provided, such as common-patterns.md and debugging-checklist.md, explicitly guide the agent to recognize and remediate security vulnerabilities (e.g., SQL Injection, XSS, insecure password storage) in the user's codebase.
  • [SAFE]: No obfuscation techniques, such as Base64 encoding of commands or hidden characters, were found in the skill's instructions or resource files.
  • [SAFE]: The skill uses standard developer tools (linting, testing, security scanners) and well-defined MCP tools for code exploration, with no evidence of remote code execution from untrusted sources.
  • [SAFE]: Indirect Prompt Injection risk is minimal as the skill processes bug reports and logs within a structured framework designed for diagnosis rather than instruction following.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 01:38 AM