- Memanto @moorcheh-ai/memantoMCP server for Memanto - persistent semantic memory for any MCP-compatible agentNot assessed yet.Runs on your machine, installed from PyPI.2,311 GitHub stars.
- IWE @iwe-org/iweMarkdown knowledge base as agent memory. Runs against the notes directory it is started in.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1,755 GitHub stars.
- Minutes @silverstein/minutesThe private, owned conversation-memory layer for AI. Record, transcribe, and search every meeting.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1,541 GitHub stars.
- Persome @intuition-lab/personal-modelLocal-first personal memory and model server for trusted MCP agents on macOS.10% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1,332 GitHub stars.
- Nautilus Compass Memory @chunxiaoxx/nautilus-compassLocal-first long-term memory for AI agents, with byte-recomputable signed verification receipts.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.1,260 GitHub stars.
- Token Savior @mibayy/token-saviorStructural codebase MCP server with persistent memory: navigate by symbol, recall across sessions.9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1,160 GitHub stars.
- Deja Vu @vshulcz/deja-vudeja-vu: local memory over the session histories of forty-one coding agents.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1,159 GitHub stars.
- Neo4j Memory @neo4j-contrib/mcp-neo4j-memoryMCP Neo4j Knowledge Graph Memory Server3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.987 GitHub stars.
- Projectmem @riponcm/projectmemCoding agent memory — one local MCP server for every project. Warns before repeating failed fixes.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.856 GitHub stars.
- Memorix @avids2/memorixLocal-first project memory with legacy MCP and 2026 discovery compatibility.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.829 GitHub stars.
- Kody @kentcdodds/kodyYour agents' cloud. MCP server with search, execute, packages, jobs, secrets, and integrations.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.753 GitHub stars.
- Vestige @samvallad33/vestigeLocal-first memory for AI agents that reaches backward to find a failure's root cause.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.649 GitHub stars.
- ThoughtDAG @chenxiachan/thoughtdagSearch local AI conversations by file or phrase and recall source turns. Four read-only tools.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.611 GitHub stars.
- Compartment @maxfreedompollard/compartmentDurable agentic memory, encrypted at rest. Fully offline: no network, no API key, no cloud.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.579 GitHub stars.
- Token Optimizer @ooples/token-optimizer-mcpContext-window optimization: caching, compression & smart file tools that cut tokens 60-90%.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.541 GitHub stars.
- Telemem @teleai-uagi/telememLong-term and multimodal memory for AI agents - character-aware, mem0-compatible, fully-local option4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.493 GitHub stars.
- GrayMatter @angelnicolasc/graymatterPersistent agent memory as a single Go binary: knowledge graph, TUI observability, zero infra.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.480 GitHub stars.
- Marm @lyellr88/marm-mcp-serverUniversal MCP Server with advanced AI memory capabilities and semantic search.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.419 GitHub stars.
- AgentRecall @goldentrii/agent-recallCorrection-first agent memory. Precision KPI tracks if agents heed warnings. 5 layers, local-only.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.372 GitHub stars.
- Ori Memory @aayoawoyemi/ori-memoryPersistent memory infrastructure for AI agents. Identity, knowledge graph, and search.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.331 GitHub stars.
- Shodh Memory @varun29ankus/shodh-memoryCognitive memory for AI agents — semantic search, Hebbian learning, knowledge graphs.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.300 GitHub stars.
AI memory MCP servers
A server here gives an AI assistant long-term memory of past conversations, facts or preferences.
Not here: note taking apps and team wikis.
428 remote795 localTagged by Odel's AI, which can make mistakes
Most popular in AI memory
By GitHub stars, at most two per publisher- Number 1: screenpipe@screenpipe/screenpipe-mcpSearch your local screen recordings, audio transcripts, and computer activity from screenpipe.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.21,883 GitHub stars.
- Number 2: Honcho@plastic-labs/honchoMemory that reasons: continual learning for stateful agents. Better context, fewer tokens.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.7,549 GitHub stars.
- Number 3: Basic Memory@basicmachines-co/basic-memoryLocal-first knowledge management with bi-directional LLM sync via Markdown files.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.4,132 GitHub stars.
