flow-nexus-neural

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions direct users to install and run the flow-nexus package using npx flow-nexus@latest. This is a standard method for deploying the vendor's MCP server and CLI tools from the public npm registry.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it is designed to ingest and process external datasets and input data for training and prediction tasks.
  • Ingestion points: Untrusted data enters the agent's context through parameters such as dataset in mcp__flow-nexus__neural_train_distributed and input or input_data in the prediction tools.
  • Boundary markers: The provided examples do not include explicit instructions or delimiters to isolate data from commands.
  • Capability inventory: The skill has the ability to deploy compute nodes, manage distributed clusters, and execute training routines within sandboxed environments.
  • Sanitization: No data sanitization or validation mechanisms are described in the setup or usage instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 06:12 AM
Security Audit — agent-trust-hub — flow-nexus-neural