Skip to main content
Glama
1msg

1msg MCP

Official
by 1msg

@1msg/mcp

MCP server for the 1msg Chat API — 60 public tools for Cursor, Claude Desktop, and other MCP clients.

Transports: stdio (local) and Streamable HTTP (hosted).

Source of truth for codegen of tools.generated.ts / handlers.generated.ts still lives in the 1msg-api monorepo (packages/mcp). This repository is the distribution package for running and hosting the server.

Quick start

npx (stdio)

export ONE_MSG_BASE_URL=https://api.stage.1msg.io
export ONE_MSG_INSTANCE_ID=your-instance-id
export ONE_MSG_TOKEN=your-channel-token

npx -y @1msg/mcp
# or, after clone + build:
# npm start

Deprecated aliases still work: CHAT_API_BASE_URL, CHAT_API_TOKEN, CHAT_API_INSTANCE_ID.

Clone and run

git clone https://github.com/1msg/1msg-mcp.git
cd 1msg-mcp
npm install
npm run build

export ONE_MSG_BASE_URL=https://api.stage.1msg.io
export ONE_MSG_INSTANCE_ID=your-instance-id
export ONE_MSG_TOKEN=your-channel-token

npm start                 # stdio → dist/index.js
npm run start:http        # Streamable HTTP → dist/http.js

Related MCP server: Vox MCP

Environment variables

Variable

Required

Description

ONE_MSG_BASE_URL

yes (stdio)

API base, e.g. https://api.stage.1msg.io or https://api.1msg.io

ONE_MSG_TOKEN

yes (stdio)

Channel API token

ONE_MSG_INSTANCE_ID

yes (stdio)

Channel instance id

MCP_HTTP_HOST

no

HTTP bind host (default 0.0.0.0 in Docker)

MCP_HTTP_PORT

no

HTTP port (default 3100)

MCP_RATE_LIMIT_RPM

no

Per-token rate limit for HTTP (default 60)

For hosted HTTP, upstream ONE_MSG_BASE_URL is set on the server; clients send credentials per request (see below).

Cursor (mcp.json)

Local stdio

{
  "mcpServers": {
    "1msg": {
      "command": "npx",
      "args": ["-y", "@1msg/mcp"],
      "env": {
        "ONE_MSG_BASE_URL": "https://api.stage.1msg.io",
        "ONE_MSG_INSTANCE_ID": "your-instance-id",
        "ONE_MSG_TOKEN": "your-channel-token"
      }
    }
  }
}

Hosted URL (Streamable HTTP)

Prod: https://mcp.1msg.io/mcp · Stage: https://mcp.stage.1msg.io/mcp

{
  "mcpServers": {
    "1msg": {
      "url": "https://mcp.1msg.io/mcp",
      "headers": {
        "Authorization": "Bearer your-channel-token",
        "X-Instance-Id": "your-instance-id"
      }
    }
  }
}

Header aliases: X-1msg-Instance-Id (and deprecated X-Chat-Api-Instance-Id). Do not commit real tokens.

Hosted HTTP locally

export ONE_MSG_BASE_URL=https://api.stage.1msg.io
export MCP_HTTP_PORT=3100
npm run start:http
  • Auth: Authorization: Bearer <token> + X-Instance-Id: <instanceId>

  • Health: GET /healthz, GET /readyz

docker build -t 1msg/1msg-mcp-http:local .
docker run --rm -p 3100:3100 \
  -e ONE_MSG_BASE_URL=https://api.stage.1msg.io \
  1msg/1msg-mcp-http:local

Scripts

Script

Purpose

npm run build

Compile TypeScript → dist/

npm test

Jest unit tests

npm run smoke:tools-list

Stdio tools/list smoke (expects 60 tools; uses dummy token)

npm start / npm run start:http

Run stdio / HTTP servers

Architecture

All Chat API HTTP goes through @1msg/sdk. See ARCHITECTURE.md.

License

MIT

Install Server
A
license - permissive license
C
quality
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • F
    license
    -
    quality
    D
    maintenance
    A command-line interface application enabling interactive chat with AI models via the Anthropic API. It supports document retrieval, command-based prompts, and extensible tool integrations through the MCP architecture.
  • A
    license
    A
    quality
    B
    maintenance
    Enables MCP clients like Claude Code and Cursor to use multiple AI models (Gemini, GPT, Grok, DeepSeek, Kimi, Ollama) via a unified chat tool with conversation memory.
    3
    Apache 2.0
  • A
    license
    -
    quality
    A
    maintenance
    Connects MCP clients (Claude Desktop, Cursor, Cline, Zed) to llmsproxy.ai, exposing chat, coding, and retrieval capabilities as MCP tools, resources, and prompts.
    8
    Apache 2.0

View all related MCP servers

Related MCP Connectors

  • Real-time chat hub for AI agents — Claude Code, Cursor, Cline, Codex over MCP or REST.

  • Free public MCP for AI agents — 193 tools, 44 workflows. No API key.

  • MCP connector that lets ChatGPT list, search, and run your Apple Shortcuts via a local Mac agent

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/1msg/1msg-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server