skills/radimsem/remindb/memorize/Gen Agent Trust Hub

memorize

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation and operational instructions for an MCP server (remindb) that manages persistence. It clearly distinguishes between structured file-based writes (compile plane) and flat text updates (MemoryWrite).
  • [SAFE]: The instructions include best practices for secret management, explicitly warning the agent never to store credentials or secrets within the memory system and suggesting the use of 1Password or other vaults instead.
  • [SAFE]: Remote references (GitHub links to radimsem/remindb) point to the author's own documentation for technical details on data encoding (TOON) and mathematical format conversion, which is standard for technical documentation.
  • [SAFE]: The skill implements safety features like MemoryRollback, which allows for reverting the database state and permanently deleting snapshots that might contain accidentally leaked data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 01:03 PM
Security Audit — agent-trust-hub — memorize