understand-jarvis

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a local file (JARVIS-FRESH/.understand-anything/knowledge-graph.json) and user-provided arguments. This creates a surface for indirect prompt injection where malicious content within the knowledge graph could attempt to influence the agent's behavior. However, the skill's current capabilities are limited to information retrieval and natural language explanation, which significantly mitigates this risk.
  • Ingestion points: Reads JARVIS-FRESH/.understand-anything/knowledge-graph.json and parses $ARGUMENTS.
  • Boundary markers: None explicitly defined to separate data from instructions.
  • Capability inventory: Performs file reads and calls the mcp__memory__search_nodes tool.
  • Sanitization: No specific sanitization or validation of the JSON content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 12:07 PM
Security Audit — agent-trust-hub — understand-jarvis