skills/h-mmer/pentest-agents/analyze/Gen Agent Trust Hub

analyze

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Python scripts (brain.py, intel_engine.py) using the uv tool and passes the $ARGUMENTS variable directly into the shell command. This pattern is susceptible to command injection if the user input contains shell metacharacters.
  • [PROMPT_INJECTION]: The skill processes untrusted data from the recon/ directory and tool outputs, creating a surface for indirect prompt injection where malicious instructions in the data could hijack the agent's execution flow.
  • Ingestion points: files within the recon/ directory, stdout from $CLAUDE_PROJECT_DIR/tools/brain.py, and stdout from $CLAUDE_PROJECT_DIR/tools/intel_engine.py.
  • Boundary markers: Absent; the skill does not use delimiters or instructions to ignore embedded commands in the processed data.
  • Capability inventory: Directory listing (ls), file reading, and shell command execution via uv run python3.
  • Sanitization: Absent; there is no evidence of input validation or output escaping before data is processed or used in subsequent logic.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 07:58 AM
Security Audit — agent-trust-hub — analyze