threat-modeling

Fail

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: CRITICALCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill automates security analysis by executing local Python scripts such as module_discovery.py for project scanning and phase_data.py for workflow management. It also recommends using established security tools like Semgrep or CodeQL if present on the system to enhance discovery.
  • [EXTERNAL_DOWNLOADS]: Installation instructions guide users to download the skill from the author's official GitHub repository (github.com/fr33d3m0n/threat-modeling).
  • [PROMPT_INJECTION]: The SKILL.md contains strict behavioral instructions requiring the AI agent to follow the defined 8-phase workflow rather than attempting manual analysis, ensuring consistent and professional security deliverables.
  • [SAFE]: Automated security alerts for 'infected' content or 'obfuscated execution chains' relate to OWASP reference files and XSS filter evasion cheat sheets within the knowledge/ directory. These are security testing references containing attack examples and are not malicious components of the skill.
Recommendations
  • CRITICAL: 1 infected file(s) detected - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Apr 26, 2026, 02:41 PM
Security Audit — agent-trust-hub — threat-modeling