MIST

MIST is an MCP server that connects AI assistants to local file-based notes, Google services, and local Git repositories. It bridges LLM clients directly to productivity and developer workflows, allowing users to query, draft, and modify information across several external systems using natural language. Designed for developers, researchers, and knowledge workers who rely on Google Workspace alongside local tools, MIST handles communication across multiple endpoints including Gmail, Google Calendar, and Google Tasks. Users can prompt their assistant to inspect email inboxes, reply to messages, create and reorganize scheduled calendar events, mark checklist items complete, and create or retrieve categorized notes stored on their local filesystem. Additionally, the server incorporates Git controls, enabling assistants to inspect local repository statuses, review diffs, checkout branches, and stage or commit code changes. By acting as a standardized interface across these different tools, MIST consolidates personal task management and software development actions into a unified assistant environment.

Category: Communication & Messaging

Tags: calendar, git, gmail, notes, tasks

Visit MIST

How to install and configure MIST

1. Ensure Python 3.13+ is installed, then clone the repository: bash git clone https://github.com/cloaky233/mist.git cd mist 2. Install dependencies using uv: bash uv install (Alternatively: pip install -e .) 3. Set up Google Cloud OAuth credentials for Gmail, Calendar, and Tasks APIs, saving the client configuration as credentials.json. 4. Create a .env file setting MIST_NOTES_DIR, MIST_GOOGLE_CREDENTIALS_PATH, and MIST_GOOGLE_TOKEN_PATH. 5. Add the server to your Claude Desktop configuration (claude_desktop_config.json): json { "mcpServers": { "M.I.S.T.": { "command": "uv", "args": [ "run", "--with", "mcp[cli]", "--with-editable", "/path/to/mist", "mcp", "run", "/path/to/mist/server.py" ], "env": { "MIST_GOOGLE_CREDENTIALS_PATH": "/path/to/mist/credentials.json", "MIST_GOOGLE_TOKEN_PATH": "/path/to/mist/token.json", "MIST_NOTES_DIR": "/path/to/notes" } } } } 6. Run the server and complete the browser-based OAuth flow on initial launch.

What you can do with MIST

• Searching, composing, and triaging messages or managing custom labels in Gmail directly through conversational prompts. - Scheduling, searching, updating, or clearing events on Google Calendar during meeting planning conversations. - Managing personal to-do lists in Google Tasks by creating entries, setting due dates, and marking finished tasks. - Creating, tagging, editing, and searching local persistent notes and summaries across project directories. - Inspecting Git status, viewing diffs, creating branches, and committing project changes through assistant commands.

Key facts

• https://github.com/CLoaKY233/MIST
• Communication & Messaging, Notes & Knowledge Management, Project & Task Management
• calendar, git, gmail, notes, tasks

Part of MCP Servers

How do I install MIST?

You install MIST by cloning the GitHub repository, installing requirements with UV or pip using Python 3.13 or newer, and setting up a Google Cloud project with OAuth credentials. You then configure the paths in an environment file and register the server inside your client configuration.

What can MIST do?

MIST allows AI assistants to manage local rich text notes, read and send emails via Gmail, schedule and update Google Calendar events, manage Google Tasks, and perform repository actions like git diff, commit, and branch management.

Which MCP clients work with MIST?

MIST works with any standard Model Context Protocol client. The documentation explicitly provides setup configurations for Claude Desktop and Zed Editor using standard UV execution commands.

Is MIST open source?

Yes, MIST is open source software distributed under the MIT license. You can inspect, modify, and contribute to the source code on GitHub.