vishalbanwari26/mnemos
by Various
Persistent memory for LLM agents — episodic/semantic/procedural memory, auditable forgetting, and a benchmark proving it actually recalls things days later.
MCP
vishalbanwari26/mnemos
Added 13 Sept 2026
Overview
Mnemos provides persistent memory for LLM agents using episodic, semantic, and procedural memory types. It includes auditable forgetting mechanisms and a benchmark that demonstrates recall over days. Written in Python.
Best for
Best for
Developers building Python-based LLM agents that need durable memory.
Use cases
- Give an LLM agent long-term recall across sessions
- Implement structured memory types for agent workflows
- Evaluate memory retention with a built-in benchmark
How to use
Tools exposed
semantic_heavyrecency_heavy
Clients listed in this record (not verified here)
Claude Code, ChatGPT
Notes
0 stars on GitHub. Last updated 2026-08-29. Licensed MIT.
Pros
- Supports three memory types for flexible storage
- Includes auditable forgetting for controlled memory decay
- Comes with a benchmark to verify recall over time
Cons
- Zero stars indicates minimal community adoption
- Unclear maintenance status or documentation depth
- Python-only, limiting language compatibility
Open-source & AI alternatives
Swap-in tools that solve the same job. Weigh the trade-offs before you commit.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.