memos-memory-guide

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a set of tools (memory_search, skill_get, skill_install, etc.) intended for managing agent state and discovering new capabilities within the MemOS ecosystem. These tools are vendor-provided resources from memtensor and serve the skill's primary purpose.
  • [PROMPT_INJECTION]: The skill inherently possesses an indirect prompt injection surface by retrieving untrusted historical data and skill guides.
  • Ingestion points: Data is ingested via the memory_search and skill_get tools which retrieve user history and public guides.
  • Boundary markers: No explicit delimiters or 'ignore-previous-instructions' warnings are specified for processing the retrieved memory content.
  • Capability inventory: The skill_install tool allows the agent to adopt behaviors discovered through external data sources.
  • Sanitization: The documentation does not outline any specific sanitization or validation mechanisms for the processed memory data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 03:55 AM