Agents IA
iwe
A markdown knowledge graph with LSP for editors and MCP for agents.
★ 1,7 k étoilesApache-2.0
Voir le guide →basic-memory est un projet de agents IA open source, créé par basicmachines-co (basicmachines-co/basic-memory sur GitHub). Il compte 4 119 étoiles sur GitHub et 305 forks, il est publié sous la licence AGPL-3.0 et a été mis à jour pour la dernière fois sur GitHub le 8 octobre 2026.
Basic Memory: local-first MCP memory so you and your AI tools build a shared knowledge graph in Markdown
Install Basic Memory with uv (the README currently requires the prerelease flag).
Add the MCP server to Claude Desktop, Codex, Cursor, VS Code, or Claude Code with the documented config.
Ask your AI to save a decision or fact; watch it appear as a Markdown file in ~/basic-memory.
Open the same files in Obsidian or your editor, edit them, and let sync merge your changes with the AI's.
Memory is human-readable Markdown files you own — no proprietary store or lock-in.
Human edits and AI writes merge continuously instead of overwriting each other.
Observations and wikilinks in notes form a navigable graph your AI can traverse.
FastEmbed embeddings with full-text + vector hybrid search and optional reranking, on SQLite or Postgres.
Infer, validate, and diff note schemas so structured knowledge stays consistent as it grows.
Claude, Codex, Cursor, ChatGPT, and other MCP clients share the same memory; Obsidian works on the same files directly.
Clone the project and enter its directory (most users install the tool, next step).
git clone https://github.com/basicmachines-co/basic-memory.git
cd basic-memoryInstall Basic Memory as a uv tool (prerelease flag currently required per the README).
uv tool install basic-memory --prerelease=allowRegister the server with your client, then restart the client and ask it to save a memory.
claude mcp add basic-memory -- uvx --prerelease=allow basic-memory mcpBasic Machines (basicmachines-co) develops Basic Memory and offers an optional hosted Cloud.
Local use is free under AGPL-3.0. The optional Cloud (sync/mobile) is a separate hosted service offered by Basic Machines.
No — local embeddings (FastEmbed) and local files need no key or account.
Those hide memory in embeddings. Basic Memory keeps it as plain Markdown you can read, edit, and version, with a knowledge graph and search layered on top.
Run `uv tool install basic-memory --prerelease=allow`, add the MCP config to your client, restart it, and ask your AI to remember something.
A markdown knowledge graph with LSP for editors and MCP for agents.
High-performance code intelligence MCP server: your codebase as a persistent knowledge graph
One portable, local-first memory layer for every AI agent - Markdown-native, user-owned and self-evolving.
Local-first AI memory that runs offline on just 8 GB of RAM.
LongMemory: local-first memory for AI agents with temporal truth, provenance, and hybrid recall
The open-source memory layer for autonomous agents — knowledge-graph memory that emulates human memory
Les autres projets de la catégorie Agents IA avec le plus d'étoiles du catalogue LibreAI — la sélection réaliste d'alternatives si basic-memory ne vous convient pas.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Projets de la catégorie Agents IA proches dans le catalogue, les plus proches en étoiles de basic-memory — continuez à explorer de part et d'autre.
Open-source, vision-first browser agent
Langroid: a principled Python framework for building LLM multi-agent applications (from CMU/UW researchers)
Excel MCP Server: let AI agents read, write, format, and chart real .xlsx workbooks — no Excel required
Rakazo — an open-source, self-hosted AI agent for persistent teammates: bring your own model and computer, run it on web, desktop, iPhone and Android
AdalFlow: The library to build & auto-optimize LLM applications.
OpenScience — the open-source AI workbench for scientific research: give it a goal, it reads the literature, runs the experiments and writes up the findings
Gratuit et ouvert : récupérez le code source, mettez une étoile au dépôt et commencez à créer.