claude-consult

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill documents a pattern for ingesting untrusted codebase content into the agent's context for analysis by specialist roles. This establishes a surface for indirect prompt injection, where malicious instructions hidden in the analyzed codebase could attempt to subvert the agent's logic.\n
  • Ingestion points: Local codebase files (e.g., src/auth/tests.py, src/api/handler.py) read by agents via the claude CLI.\n
  • Boundary markers: The documentation does not specify the use of delimiters or instructions to 'ignore' embedded commands in the files being analyzed.\n
  • Capability inventory: Specialist agents can read and analyze repository structure, explain code logic, and perform security audits.\n
  • Sanitization: No sanitization or validation of the processed codebase content is mentioned in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 06:52 PM
Security Audit — agent-trust-hub — claude-consult