flow-nexus-neural
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs users to install and run the 'flow-nexus' package using npx. This initiates the download and execution of external code from a non-standard source to enable the skill's capabilities.
- [COMMAND_EXECUTION]: The prerequisites section details bash commands for initializing the Flow Nexus MCP server and performing user authentication.
- [DATA_EXFILTRATION]: The skill transmits model configurations, dataset identifiers, and inputs to 'flow-nexus.ruv.io'. Users should ensure they trust this external service with their metadata and processing tasks.
- [PROMPT_INJECTION]: Ingesting external data from the model marketplace creates a surface for indirect prompt injection. Maliciously crafted template metadata could potentially influence the agent's behavior.
- Ingestion points: Template list and metadata from 'mcp__flow-nexus__neural_list_templates'.
- Boundary markers: None identified.
- Capability inventory: Remote training, prediction, and cluster node deployment.
- Sanitization: No sanitization of external template content is specified.
Audit Metadata