mcp server

mycontext

A local environment to collect, manage, and own your personal data. Includes an MCP server for AI access and a simple web UI for data capture.

bencoleman24/mycontextcommunitylistedunreviewed

Use tool

Set up

Runs on your computer. Pick your agent and operating system.

Before you start

Node.js 20 or newer

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "mycontext": {
      "command": "npx",
      "args": [
        "-y",
        "mycontext"
      ]
    }
  }
}

same format Cursor, GitHub Copilot CLI, Junie, JetBrains AI Assistant, Augment (Auggie CLI), Factory Droid, Kiro, Devin CLI +12

Values in angle brackets are placeholders. Restart Claude Desktop after editing its config.Claude Desktop docs

README

synced Oct 21, 2026

mycontext

Own the context your AI assistant reads.

A local environment to collect, manage, and own your personal data.

Includes:

a local workflow for personal data management
an MCP server, so AI apps like ChatGPT, Claude, Cursor, and Gemini can read and write the data directly
plain JSON files you can read without this app, and export as Markdown, JSON, or CSV for personal use

Why mycontext

Most AI assistants today keep their memory of you on their own servers. That makes it harder to switch providers, and gives them a lot of control over your data. I think this issue will be exacerbated as AI becomes more integrated with our day to day life. I think it's worth keeping/maintaining a personal collection of data that you fully own.

mycontext keeps that data in plain files on your machine and this project contains tools to help you manage and utilize this data.

Works with

AI apps that can run MCP servers on your computer, including the ChatGPT desktop app, Claude Code, Claude Desktop, Codex CLI, Cursor, Gemini CLI, and VS Code with GitHub Copilot. See Connect an app.

Web and mobile chat apps can't reach a server on your computer. For those, export your data from the web UI and paste it into the chat.

Quickstart

git clone https://github.com/bencoleman24/mycontext.git
cd mycontext
npm install
npm run setup

npm run setup builds the project, then prints instructions for connecting mycontext to the AI apps it finds on your computer. It doesn't change any app's settings. Instructions for every supported app are in Connect an app.

Then start the web UI:

npm run web

It opens at http://127.0.0.1:4823.

Documentation


Web UI	The web app
MCP tools and resources	What AI apps can read and write, and how to connect them
Data model	Files on disk, habit rules, schema versions, configuration
Journal summaries	Summarizing entries to keep exports small
HTTP API	Routes for scripting
Architecture	How the code is organized

Privacy

Your data stays on your machine. No telemetry, no cloud sync, no external network calls. The web UI binds to local host only.

Two files in the codebase can make network calls:

src/lib/summarizer.ts — calls Ollama on localhost, ONLY when you ask for a summary
src/web/public/app.js — the web UI calling its own backend on 127.0.0.1

test/noExternalNetworkCalls.test.ts checks this. If any other file starts making network calls, the build fails.

License

MIT