A Model Context Protocol Server for Pica, built in TypeScript.

Setup Video: https://youtu.be/JJ62NUEkKAs

Demo Video: https://youtu.be/0jeasO20PyM

Model Context Protocol (MCP) is a system that lets AI apps, like Claude Desktop, connect to external tools and data sources. It gives a clear and safe way for AI assistants to work with local services and APIs while keeping the user in control.

Pica is a powerful agentic tooling platform that enables connections to 70+ third-party services and applications. This MCP server allows Claude Desktop and anything using the Model Context Protocol to securely interact with all these connections through natural language requests.

With Pica MCP Server, you can:

• Access Multiple Data Sources: Query databases, fetch files, and retrieve information across services
• Automate Workflows: Trigger actions and automate tasks across your connected platforms
• Enhance LLM Capabilities: Give Claude Desktop powerful real-world capabilities through API access

Pica supports 70+ connections (with more added regularly) across categories like:

• Gmail, Outlook Mail, Slack, Teams, SendGrid
• Notion, Google Drive, Dropbox, OneDrive

• PostgreSQL, BigQuery, Supabase
• Google Sheets, Airtable

• Salesforce, HubSpot, Pipedrive, Zoho
• Zendesk, Freshdesk, Intercom

• OpenAI, Anthropic, Gemini, ElevenLabs

• Shopify, BigCommerce, Square, Stripe
• QuickBooks, Xero, NetSuite

You can use this package directly with npx:

npx @picahq/pica-mcp

Or install it globally:

npm install -g @picahq/pica-mcp

To install pica for Claude Desktop automatically via Smithery:

npx -y @smithery/cli install @picahq/pica --client claude

This server requires a Pica API key. Set the environment variable:

export PICA_SECRET=your_pica_secret_key

Build the Docker Image:

docker build -t pica-mcp-server .

Run the Docker Container:

docker run -e PICA_SECRET=your_pica_secret_key pica-mcp-server

Install dependencies:

npm install

Build the server:

npm run build

For development with auto-rebuild:

npm run watch

To use with Claude Desktop, add the server config:

On MacOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json

On Windows: %APPDATA%/Claude/claude_desktop_config.json

To use the Docker container with Claude Desktop, update your claude_desktop_config.json with:

{
  "mcpServers": {
    "pica-mcp-server": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "-e", "PICA_SECRET=YOUR_PICA_SECRET_KEY",
        "pica-mcp-server"
      ]
    }
  }
}

{
  "mcpServers": {
    "pica-mcp-server": {
      "command": "node",
      "args": [
        "/path/to/pica-mcp-server/build/index.js"
      ],
      "env": {
        "PICA_SECRET": "YOUR_PICA_SECRET_KEY"
      }
    }
  }
}

Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector, which is available as a package script:

npm run inspector

The Inspector will provide a URL to access debugging tools in your browser.

Once you've added the server config and connected some platforms in the Pica dashboard, restart Claude Desktop and try out some examples:

• Send an email using Gmail to a colleague with a meeting summary
• Create a calendar event in Google Calendar for next Tuesday at 2pm
• Send a message in Slack to the #marketing channel with the latest campaign metrics
• Find documents in Google Drive related to Q3 planning

• List the top 10 customers from my PostgreSQL database
• Create a new sheet in Google Sheets with sales data
• Query Salesforce for opportunities closing this month
• Update a Notion database with project statuses

• Create a support ticket in Zendesk from customer feedback
• Process a refund for a customer order in Stripe
• Add a new lead to HubSpot from a website inquiry
• Generate an invoice in QuickBooks for a client project

• Generate an image with DALL-E based on product specifications
• Transcribe a meeting recording with ElevenLabs
• Research market trends using Tavily or SerpApi
• Analyze customer sentiment from support tickets

Got any cool examples? Open a PR and share them!

This project is licensed under the GPL-3.0 license. See the LICENSE file for details.