vulnerability-validation

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill provides instructions for the AI to perform reasoning on local security findings and does not include executable code, network requests, or sensitive file access.
  • [PROMPT_INJECTION]: No attempts to override system prompts or bypass safety filters were found in the instructions.
  • [DATA_EXFILTRATION]: The skill interacts with local artifacts in the .bug-hunter/ directory which is consistent with its stated purpose. There are no patterns suggesting the exfiltration of this data to external servers.
  • [REMOTE_CODE_EXECUTION]: No package installations or remote script executions are present in the skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 01:18 AM
Security Audit — agent-trust-hub — vulnerability-validation