mnemosyne
Active·★ 151·MIT·Updated 2026-10-21
Mnemosyne provides persistent memory for AI agents, enabling them to carry work forward, recover original evidence, and adapt by retiring obsolete lessons when their source changes. It connects memory to its consequences, allowing agents to resume from handoffs, withhold skills after failures, and benefit from a wiser start.
#AI Agents#Persistent Memory#Context Management#Knowledge Base#Semantic Retrieval#SQLite#TypeScript#Multi-Agent Memory
01
Features
01Memory with evidence: Private-by-default records, explicit sharing, corrections, conflicts, source history, and dependent retirement.
02Continuity across agents: Structured checkpoints preserve goals, decisions, artifacts, and next actions; source changes can retire obsolete handoffs.
03Adaptive context: Budgeted observations, hierarchical summaries, and expandable original source bytes, with evidence revalidation.
04Recall across words and meaning: Scoped BM25 by default; optional embeddings, hybrid fusion, and reranking over authorized records.
05Evidence-gated skills: Candidate procedures, explicit trial requirements, and retirement after failed or changed evidence.
02
Compatibility
Node.js
Runtime
Verified via docs
TypeScript
SDK
Verified via docs
Python
Client
Verified via docs
SQLite
Database
Verified via docs
MCP
Client Protocol
Verified via docs
HTTP
API
Verified via docs
03
Quick start
1
$ git clone --branch v2.0.0-rc.9 --depth 1 https://github.com/28naem-del/mnemosyne.git
2
$ cd mnemosyne
3
$ npm ci --ignore-scripts
4
$ npm run build
04
Use cases
↳Building AI agents that require long-term, verifiable memory for decision-making and learning.
↳Enabling agents to resume complex tasks or projects across multiple sessions or handoffs without losing context.
↳Developing AI systems that adapt to new information by invalidating or correcting outdated knowledge based on source changes.
↳Creating self-improving agents where skills are refined or retired based on trial outcomes and evidence.
↳Managing dynamic context for large language models, ensuring relevance and adherence to evidence within token budgets.
05
Alternatives
okf-agent-memory★ 827
Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.
mex★ 1.8k
Persistent project memory for AI coding agents. Structured scaffold + drift detection CLI.
mcp-local-memory★ 60
A lightweight, powerful local memory server for AI agents supporting text, entities, and relations. Enables persistent codebase understanding and user preference management.
Waggle-mcp★ 44
MCP server for external memory layer for AI agents + more . Download from pypi , and get started
state-memory-mcp★ 48
Persistent, branch-aware workflow state memory MCP server for AI coding assistants. Tracks tasks, accepted decisions, and active blockers to prevent session context bloat and speed up development.
mori★ 32
mori MCP server — A shared memory layer for AI coding agents — one that compounds.
Comments
Log in to leave a comment
No comments yet. Be the first!