hermes-memory

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a clear and well-structured process for managing 'durable memory', focusing on architecture decisions, naming conventions, and reusable commands.
  • [SAFE]: Explicit security guidelines are provided in the documentation, specifically instructing the agent not to store sensitive secrets or tokens in memory files.
  • [SAFE]: The use of filesystem and shell tools is contextually appropriate for the tasks of indexing, reading, and writing project-specific documentation within a local repository.
  • [SAFE]: The prefetch and sync procedures incorporate summarization and filtering steps, which help maintain context quality and provide a controlled way to process local data.
  • [SAFE]: No evidence of prompt injection, data exfiltration, obfuscation, or unauthorized persistence mechanisms was found in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 05:23 PM
Security Audit — agent-trust-hub — hermes-memory