Red Team Tools and Methodology

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides numerous examples and a bash script for executing security tools such as Amass, Subfinder, httpx, and Nuclei. These are used for legitimate security research purposes.
  • [EXTERNAL_DOWNLOADS]: Includes references to established security research platforms and data sources like bgp.he.net for information gathering.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a reconnaissance workflow that ingests data from external targets via tools like subfinder, waybackurls, and gau (ingestion points in SKILL.md). It lacks explicit boundary markers or sanitization within the provided recon.sh script, while maintaining capabilities for subprocess execution and network operations (capability inventory in SKILL.md). This creates a surface where malicious target data could potentially influence the agent's logic, though this is typical for security-oriented tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 03:30 PM