zhitongblog/solomd

A markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.

What it solves

SoloMD is a local-first Markdown editor designed to be an environment where AI agents can live and operate directly on your notes. It solves the problem of fragmented AI workflows by integrating a first-class agent surface, an MCP (Model Context Protocol) endpoint, and a version-controlled vault, allowing users to chat with their notes and automate tasks via scheduled recipes without sacrificing privacy or data ownership.

How it works

SoloMD operates on a folder of Markdown files. It combines three core components:

  • The Editor: A WYSIWYG editor with features like semantic search, wikilinks, and tldraw whiteboards.
  • The Endpoint: A bundled solomd-mcp binary that exposes the vault to any MCP-compatible AI client (like Claude Desktop or Cursor) using 13 built-in tools.
  • The Agent Surface: An integrated panel for streamed chat-with-vault and "recipes"—YAML-defined jobs (e.g., weekly reviews, TODO extraction) that run on a separate AutoGit branch. This ensures that AI-generated changes are sandboxed and must be accepted or rejected by the user before merging into the main branch.

Who it’s for

It is for users who want a powerful, private, local-first knowledge base that can be actively managed by AI agents, developers who use MCP clients, and writers who prefer Markdown-based workflows with built-in versioning and AI assistance.

Highlights

  • Agent-Native Workflow: Integrated agent panel and scheduled recipes with a sandbox (AutoGit) for safe AI writes.
  • MCP Integration: Bundled MCP server allowing external LLM clients to drive the same vault.
  • MCP Federation: Ability to handle multiple vaults in a single MCP session.
  • Local-First & Private: All notes, API keys (stored in OS keychain), and embeddings remain on the machine; no cloud servers required.
  • Knowledge Graph Layer: Properties inspector for YAML frontmatter, typed relationships, and a relationship graph.
  • Broad AI Provider Support: Support for 14 BYOK (Bring Your Own Key) providers, including local models via Ollama.
  • AutoGit Versioning: Automatic local commits for every save, ensuring a complete history per note.

Related

  • Project
    marswaveai/ColaMD

    A lightweight WYSIWYG Markdown editor that syncs in real time with changes made by AI agents, providing a visual indicator of AI activity.

  • Project
    flyhunterl/flymd

    A lightweight, local-first Markdown editor featuring AI-assisted writing, RAG-based semantic search, and high-precision PDF parsing.

  • Project
    markrahq/markra

    Markra is an open‑source, local‑first Markdown editor that combines true WYSIWYG editing with built‑in AI assistance. Users can invoke LLMs (any OpenAI‑compatible or local model) to polish, rewrite, summarize, translate, or otherwise edit selected text or whole documents, with a preview step before changes are applied. The desktop app (Tauri) runs on macOS, Windows, and Linux; a web version mirrors the same features. Files stay as plain `.md` on disk, with optional WebDAV sync, image storage options, spellcheck, custom CSS themes, and export to HTML/PDF. Licensed under AGPL‑3.0 and community‑driven.

  • Project
    blueberrycongee/Lumina-Note

    A local-first Markdown workspace with an integrated AI agent that can read, write, and reorganize notes across a user's entire vault.

  • Project
    inkeep/open-knowledge

    OpenKnowledge is a free, open‑source markdown editor that feels like Google Docs/Notion but runs locally. It offers true WYSIWYG editing, desktop (macOS/Windows/Linux) and web UI, and side‑by‑side AI assistance from Claude, Codex, Pi and other LLMs via an MCP/CLI bridge. Built‑in git/GitHub sync lets teams share notes, and the tool supports AI‑driven wikis, second‑brain knowledge graphs, and embeddable HTML specs.