debugging-wizard

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and reference files were thoroughly analyzed across all threat categories. No instances of prompt injection, data exfiltration, obfuscation, or malicious command execution were found.
  • [PROMPT_INJECTION]: Instructions focus on established debugging methodologies and constraints intended to improve the quality of technical analysis without attempting to bypass safety protocols.
  • [DATA_EXFILTRATION]: No network operations or sensitive file access patterns were detected. The skill operates on provided error messages and logs within the local context.
  • [REMOTE_CODE_EXECUTION]: All referenced tools (pdb, delve, node inspector) are standard development utilities. No remote scripts or unverified third-party dependencies are downloaded or executed.
  • [COMMAND_EXECUTION]: The shell commands provided in the references are standard debugging and git operations intended for development environments.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 01:22 AM
Security Audit — agent-trust-hub — debugging-wizard