
Long-term memory layer for AI agents that stores typed facts and consolidates beliefs
- Open source
- Open-source agent
- 61.7K GitHub stars
- Knowledge and memory

Long-term memory layer for AI agents that stores typed facts and consolidates beliefs

Open-source shared memory hub that gives agent teams common conversation and code memory
Local-first Markdown vault memory and semantic search for AI coding assistants
Back up agent memory, files and dashboards, then recover from a reviewed snapshot
Not checked yet
Shared rules, skills and decisions for teams, served to AI agents with human review
Shared notes that language models leave for later instances, readable over MCP or GET
Agent memory of company rules a person confirmed, with source and date
A hosted outline that Claude or ChatGPT can read, write and remember in over MCP
Hosted persistent memory for AI agents, with hybrid keyword and vector search
Not checked yet
Search your markdown notes and review captured agent chats before keeping them
Not checked yetFree public home where AI agents register, keep records, read an inbox and chat in rooms
Give your AI agent a personal finance workspace with goals, a budget and account reviews

One MCP connection for your servers, skills, agents and memory across devices

Remote MCP server that lets agents read, grow and act on your BlackOps knowledge bases
Not checked yetMarketplace where agents buy and sell sealed research trails, paid in USDC via x402
Not checked yetHosted agent memory that learns which memories helped, with shared rooms across tools
Not checked yetShared notebook, run log and handoff notes that let AI agents pass work across apps
Not checked yetAgent plans and decisions drawn as a map you can read, correct and share
Not checked yetHosted agent memory with provenance, version history, snapshots and rollback
Not checked yet
Persistent document knowledge for agents with hybrid search, reranking and GraphRAG
Not checked yetSigned, offline-checkable records of places, files and observations for AI agents
Not checked yetPersistent memory and a tamper-evident audit log for agents, with causal lineage
Runs locally
Local-first shared memory layer that any MCP client can save to and recall from
Runs locallyShared memory that Claude, ChatGPT and Cursor save to and your team can read
Not checked yetA shared fact layer: sourced, time-aware claims that agents check before researching
Not checked yetShared business memory for agents: free fit assessment and a one-hour sandbox
Not checked yetTeam decisions, goals and constraints from Engramic for any MCP-capable agent
Not checked yetStorage, search, inbound email and scheduled requests for agents under a hard spend cap
Not checked yetA shared record of people, projects and events that every agent you use can read and write
Not checked yetCross-chat memory for agents with a draft, check and seal step before answers
Not checked yet