iwe
A local-first hierarchical knowledge graph in Markdown that serves as shared memory for you and your AI agents.
What is iwe?
IWE is a memory system that organizes personal and project knowledge into a hierarchical graph using Markdown files. It serves as a shared second brain for users and AI agents, providing structured access to notes so AI can retrieve and update information with full context. Unlike opaque vector databases, IWE keeps everything in readable, editable Markdown, ensuring users remain in control.
Developers, researchers, and knowledge workers who manage extensive notes and want to integrate AI agents into their workflow. It's ideal for those using AI tools like Claude, Codex, or Gemini and needing consistent, structured context.
Developers choose IWE because it offers a transparent, local-first knowledge graph that both humans and AI can navigate seamlessly. Its unique selling point is the combination of hierarchical Markdown organization with powerful AI integration via CLI and MCP, avoiding the black-box nature of embedding-based systems.
Overview
Markdown memory system for you and your AI agent
Use Cases
Best For
- Creating a shared knowledge base for AI agents to access during coding tasks
- Managing hierarchical project documentation with cross-references
- Building a personal second brain with editor-integrated refactoring
- Providing structured context to AI tools via the Model Context Protocol
- Organizing research notes that need to appear in multiple contexts
- Scripting automated knowledge workflows with a command-line interface
Not Ideal For
- Teams requiring real-time cloud collaboration without manual sync setup
- Non-technical users preferring graphical, WYSIWYG editors over Markdown and CLI
- Projects needing built-in AI features like automatic summarization or content generation
- Simple note-taking without hierarchical or AI integration needs
Pros & Cons
Pros
Supports polyhierarchy with inclusion links, allowing notes to appear under multiple parents without duplication, enabling rich context inheritance as per the concepts documentation.
Provides CLI and MCP server for AI agents to retrieve, update, and navigate knowledge with full context, making it ideal for structured collaboration with external AI tools like Claude.
Offers IDE-like features via LSP in VS Code, Neovim, Zed, and Helix, including search, navigation, refactoring, and auto-complete, enhancing Markdown productivity.
Keeps all data as plain Markdown files on your machine, ensuring data control and privacy without reliance on cloud services or opaque databases.
Built in Rust for instant processing of thousands of notes, making it efficient for large knowledge bases with blazing fast operations.
Cons
Relies entirely on external AI tools for intelligent features, requiring users to set up and manage separate AI services, which adds complexity.
Requires installation of CLI, LSP server, and editor configuration via tools like Cargo or Homebrew, which may be daunting for non-developers.
Lacks native real-time sync or collaborative editing features; teams must rely on external version control systems like Git for sharing, which can be cumbersome.
Frequently Asked Questions
Related Projects
cc-switchA cross-platform desktop All-in-One assistant tool for Claude Code, Codex, OpenCode, openclaw & Gemini CLI.
ScreenpipeRun agents that work based on what you do. 24/7 local screen & mic recording for the superintelligence era
espansoA Privacy-first, Cross-platform Text Expander written in Rust
ast-grep⚡A CLI tool for code structural search, lint and rewriting. Written in Rust
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.