IWE

Active
GitHub Rust Apache-2.0

Description

Markdown memory system for you and your AI agent. Provides persistent memory storage through structured Markdown files with context management and retrieval.

Key Features

  • Markdown knowledge graph: transforms a local directory of Markdown files into a queryable knowledge graph
  • MCP server integration: lets AI agents search, retrieve, and refactor notes via the Model Context Protocol
  • Editor LSP support: native integration with VS Code, Neovim, Zed, Helix — jump-to-definition, search, and autocomplete
  • Built in Rust for speed: processes 20,000 files in under a second
  • Fully local, zero lock-in: Markdown files are the data — no cloud, no database, no vendor lock-in
  • Hierarchical linking system: supports nesting and cross-references, one note can belong to multiple topics

Use Cases

💡 Personal knowledge management: build a queryable personal knowledge base from Markdown files
💡 AI agent memory: provide structured long-term memory and context retrieval for AI agents
💡 Technical documentation management: build a searchable, refactorable technical documentation knowledge graph
💡 Team collaboration: share and collaboratively edit Markdown knowledge bases via git version control

Tags

rust memory agent tools mcp

Categories

🧩 Agent Memory

Quick Start

1. Install CLI: `brew tap iwe-org/iwe && brew install iwe` (or `cargo install iwe iwes iwec`);
2. Initialize workspace: `cd ~/notes && iwe init`;
3. Set up editor: install the IWE extension for VS Code/Neovim/Zed/Helix;
4. Connect AI agent: add `{"mcpServers": {"iwe": {"command": "iwec", "args": ["--project", "~/notes"]}}}` to your MCP config;
5. Use `iwe find`, `iwe retrieve`, `iwe tree` and other commands to operate on notes.
Visit GitHub Visit Website

Related Projects