nats-hub
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates communication with external agents and human operators via a messaging bus, creating an attack surface for indirect prompt injection.
- Ingestion points: Tools like get_history, get_thread, list_pending, and session management tools ingest data from the NATS bus into the agent's context.
- Boundary markers: The instructions do not define delimiters or warnings to treat bus data as untrusted.
- Capability inventory: The agent using these tools potentially has access to file systems, shell commands, or other sensitive MCP tools.
- Sanitization: No evidence of input validation or sanitization of messages from the hub is present in the instructions.
Audit Metadata