skills/mukul975/anthropic-cybersecurity-skills/performing-physical-intrusion-assessment/Gen Agent Trust Hub
performing-physical-intrusion-assessment
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation and utility scripts for physical security auditing. All Python scripts use only the standard library for checklist management and report generation, with no network or dynamic code execution capabilities.
- [SAFE]: The documentation in SKILL.md correctly classifies offensive techniques (e.g., badge cloning, lock bypassing) and provides example configurations for hardware tools. These examples include placeholder C2 server addresses (c2.redteam.com) and stager payloads, which are standard for red-teaming documentation and do not pose a risk to the host system.
- [SAFE]: Analysis of scripts/agent.py and scripts/process.py confirms they are legitimate data processing tools for auditing purposes, with no evidence of credential harvesting, unauthorized file access, or persistence mechanisms.
Audit Metadata