agents-os
Collection of MCP servers for AI agent teams - Seamless integrations with Airtable, Asana, and Slack for powerful AI-driven workflows
claude mcp add --transport stdio jakreymyers-agents-os node servers/asana-mcp-server/dist/server.js \ --env ASANA_ACCESS_TOKEN="Your Asana access token"
How to use
Agents OS presents three MCP servers that bridge AI agents with external tools: Airtable for database operations, Asana for project and task management, and Slack for workplace collaboration. Each server exposes MCP-compatible endpoints to perform common actions such as listing and querying records, creating and updating tasks or projects, and interacting with Slack channels and messages. When running Claude Code or another AI agent with MCP enabled, you can compose requests that reference these services to read data, write updates, or trigger workflows inside your productivity stack. The built-in environment-based token management helps keep credentials secure while providing masked output in logs.
How to install
Prerequisites:
- Node.js v18 or later
- Git
- Internet access to clone submodules and install dependencies
Steps:
- Clone the repository:
git clone https://github.com/jakreymyers/agents-os.git
cd agents-os
- Initialize and update submodules (each server lives under servers/):
git submodule update --init --recursive
- Install dependencies for all MCP servers and build:
cd servers/airtable-mcp-server
npm install
npm run build
cd ../../asana-mcp-server
npm install
npm run build
cd ../../slack-mcp-server
npm install
npm run build
- Prepare environment variables. Copy the example template and populate your tokens:
cp .env.example .env.local
Then edit .env.local to include AIRTABLE_TOKEN, ASANA_ACCESS_TOKEN, SLACK_BOT_TOKEN, SLACK_CLIENT_ID, SLACK_CLIENT_SECRET, SLACK_REDIRECT_URI, SLACK_TEAM_ID, etc. 5) Run the servers. If you have built dist/server.js files as assumed, start them individually or via a runner script if provided:
node servers/airtable-mcp-server/dist/server.js &
node servers/asana-mcp-server/dist/server.js &
node servers/slack-mcp-server/dist/server.js &
- Connect Claude Code or your MCP client and enable the MCP servers you started.
Additional notes
Notes and tips:
- The project uses git submodules for each MCP server; ensure submodules are initialized to access server code.
- Tokens should never be committed; use .env.local and add it to .gitignore.
- If a server fails to start due to missing dist files, ensure you ran the build step in that server directory and that the dist/server.js path matches your build output.
- Slack integration includes OAuth settings; you may need to complete an OAuth flow to obtain valid tokens and permissions in SLACK_* env vars.
- You can extend or customize individual MCP servers by editing their subdirectories under servers/ and rebuilding as needed.
Related MCP Servers
mcp-batchit
🚀 MCP aggregator for batching multiple tool calls into a single request. Reduces overhead, saves tokens, and simplifies complex operations in AI agent workflows.
mcp-ssh
🔐 SSH MCP Tool - AI-powered SSH management through MCP protocol | 基于MCP协议的SSH工具,为AI提供SSH远程操作能力
architect
A powerful, self-extending MCP server for dynamic AI tool orchestration. Features sandboxed JS execution, capability-based security, automated rate limiting, marketplace integration, and a built-in monitoring dashboard. Built for the Model Context Protocol (MCP).
IoT-Edge
MCP server for Industrial IoT, SCADA and PLC systems. Unifies MQTT sensors, Modbus devices and industrial equipment into a single AI-orchestrable API. Features real-time monitoring, alarms, time-series storage and actuator control.
FocusRelayMCP
Talk to your OmniFocus tasks. An OmniFocus MCP server that lets AI assistants query your tasks, projects, and tags using natural language—no more clicking through endless lists.
cadre-ai
Your AI agent squad for Claude Code. 17 specialized agents, persistent memory, desktop automation, and a common sense engine.