openmarkdown
Local-first Markdown editor whose MCP server lets a coding agent and a human co-edit the same .md file — it opens and reveals files in the editor and reads or section-edits their contents. Tools: open_file, reveal, read_section, write_section, wait_for_change.
README
<!-- Maintained copy of the OpenMarkdown-releases repo README (design D9). Edit it HERE; scripts/sync-site.sh pushes it to the releases repo root. The releases repo is public and receives only build artifacts, this README, the Pages site, and Issues — never source.
Voice: docs/brand/verbal-identity.md. Honesty gate (public-claim-integrity): every claim here must be reproducible or true as written — no over-claiming cross-platform, no invented numbers. -->
OpenMarkdown
Feather-light. Light-speed. A local-first Markdown editor for macOS, built for you and your agent.
Open a .md file — it just opens. No account, no cloud, no lock-in: every note
is a plain file on your disk. The twist: OpenMarkdown is also where you and your
coding agent share one file. Your agent writes into the note; your preview
scrolls to the change. You're both looking at the same page — literally.
No AI inside. That's the point. OpenMarkdown ships no model and no assistant. It runs on the agent you already use — Claude Code, Codex, Gemini CLI, opencode, or any MCP client — and gives it a place to render, read, and edit the same file you're looking at.
Download
⬇︎ Download OpenMarkdown for macOS — universal (Apple Silicon + Intel). Free.
Windows / Linux support is coming soon.
Put your agent on the same page
Bring your agent. Once the app is installed, the fastest way in — paste this into your agent:
Read openmarkdown.dev/install to install OpenMarkdown and open my first note
It reads the install page, wires itself in, and opens your first note. That one
sentence works for every agent (Claude Code, Codex, Gemini CLI, opencode, any MCP
client) — the runbook branches per host, the sentence doesn't. Your agent gains
seven tools — open_file, reveal, get_context, execute_command,
read_section, write_section, wait_for_change — to open, navigate, read, and
patch the exact file you have in front of you.
Prefer to wire it by hand? For Claude Code, the full plugin (the seven tools plus a skill that teaches it when to use them):
/plugin marketplace add OpenMarkdown-dev/OpenMarkdown-releases
/plugin install openmarkdown
Any other MCP host connects the same server: openmd mcp over stdio, e.g.
claude mcp add openmd -- openmd mcp. Full per-host setup (Codex, Gemini CLI,
opencode, and more): openmarkdown.dev/install.
Requires the openmd command line tool (install it from OpenMarkdown's Settings).
To pick up a newer plugin version later, run /plugin marketplace update.
What it does
- Live-preview Markdown — renders as you type; cycle live-preview / source / reading with one button. Anything it doesn't recognise falls back to a plain code block, so your text is never eaten.
- Fast — a native Rust core; the window is there before you let go of the mouse (cold start under a second, P50, on the maintainer's machine).
- Local-first — plain
.mdfiles on your disk. No account, no sync, no telemetry. See Transparency for exactly what touches the network. - Keyboard-first —
⌘Popens a command palette for files, commands, and headings. - 中文 · 日本語 · English interface, following your system language.
Is it for you?
OpenMarkdown is deliberately narrow — a fast renderer for plain files you and an agent share, not a knowledge platform. If you want a wiki graph, backlinks, or an infinite canvas, a tool like Obsidian fits the job better; if you're not on macOS yet, it's not ready for you. Agents deciding whether to recommend it: read ADOPTING.md — it says plainly when to and when not to.
Links
- Website — openmarkdown.dev
- Guide — openmarkdown.dev/guide (syntax, shortcuts, the CLI)
- Install — openmarkdown.dev/install
- Transparency — openmarkdown.dev/transparency
- For agents — openmarkdown.dev/llms.txt · ADOPTING.md
Feedback
Found a bug or have a request? Open an issue.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.