fortytwo-mcp

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the eth-account and web3 Python packages, which are standard, well-known libraries used for blockchain interactions.\n- [COMMAND_EXECUTION]: The skill instructions and documentation require the user to execute local Python scripts (scripts/fortytwo_query.py and scripts/preflight.py) and suggest curl commands for streaming output to interact with the vendor's service.\n- [INDIRECT_PROMPT_INJECTION]: The skill acts as a proxy for an external multi-agent inference service, creating a surface for indirect prompt injection.\n
  • Ingestion points: External data is received from the Fortytwo Prime API gateway in scripts/fortytwo_query.py.\n
  • Boundary markers: The content returned by the service is printed directly to the output without protective delimiters or boundary markers to prevent the agent from misinterpreting instructions within the response.\n
  • Capability inventory: The skill context allows for network operations to the vendor's gateway and blockchain RPCs, and the execution of shell scripts.\n
  • Sanitization: No validation or sanitization is performed on the text returned by the inference service before it is presented to the user or agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 01:58 PM
Security Audit — agent-trust-hub — fortytwo-mcp