graph
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill performs a 'Deep Scan' of department silos to build a knowledge graph used for RAG (Retrieval-Augmented Generation). This creates a vulnerability where malicious instructions embedded in processed documentation could influence the agent's behavior during future missions.
- Ingestion points: Files and documentation stored within the '8 Department Silos'.
- Boundary markers: None identified; the skill does not specify delimiters or instructions for the agent to ignore embedded commands in the scanned data.
- Capability inventory: The skill uses an internal tool to perform repository-wide scans and write operations (exporting
graph.json). - Sanitization: There is no mention of sanitization or filtering of the content ingested from the silos before it is indexed into the knowledge graph.
Audit Metadata