deepagents-memory
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a documentation resource providing TypeScript examples for managing agent state and long-term memory backends using standard libraries.
- [INDIRECT_PROMPT_INJECTION]: The architecture demonstrated allows agents to process data stored in a persistent backend. 1. Ingestion points: The agent reads data from the /memories/ path as shown in Example 1 and Example 3. 2. Boundary markers: The examples do not explicitly demonstrate the use of delimiters for retrieved content. 3. Capability inventory: The implementation includes file writing to the persistent store and tool-based retrieval of user preferences. 4. Sanitization: No specific data sanitization logic is included in the architectural snippets.
Audit Metadata