Fireberry CRM MCP Server

A powerful Model Context Protocol (MCP) server for seamless AI-CRM integration

Connect your AI assistants directly to Fireberry CRM with secure, real-time access to your customer data. Perform complex CRM operations through natural language interactions.

Quick Start

1. Get Your API Token

Generate your Fireberry API token following the authentication guide.

2. Install & Configure

Choose your preferred runtime:

Node.js (Recommended)

Add to your MCP configuration file:

{
    "mcpServers": {
        "fireberry-crm": {
            "command": "npx",
            "args": ["-y", "@fireberry/mcp-server@latest"],
            "env": {
                "FIREBERRY_TOKEN_ID": "<your-token-here>"
            }
        }
    }
}

Bun

{
    "mcpServers": {
        "fireberry-crm": {
            "command": "bunx",
            "args": ["@fireberry/mcp-server@latest"],
            "env": {
                "FIREBERRY_TOKEN_ID": "<your-token-here>"
            }
        }
    }
}

3. Tool-Specific Setup

Update claude_desktop_config.json from MCP official docs:

{
    "mcpServers": {
        "fireberry-crm": {
            "command": "npx",
            "args": ["-y", "@fireberry/mcp-server@latest"],
            "env": {
                "FIREBERRY_TOKEN_ID": "<your-token-here>"
            }
        }
    }
}

Add to .vscode/settings.json:

{
    "github.copilot.advanced": {
        "mcpServers": {
            "fireberry-crm": {
                "command": "npx",
                "args": ["-y", "@fireberry/mcp-server@latest"],
                "env": {
                    "FIREBERRY_TOKEN_ID": "<your-token-here>"
                }
            }
        }
    }
}

Navigate to Settings → MCP Servers and add:

{
    "fireberry-crm": {
        "command": "npx",
        "args": ["-y", "@fireberry/mcp-server@latest"],
        "env": {
            "FIREBERRY_TOKEN_ID": "<your-token-here>"
        }
    }
}

Features

🔍 Metadata & Discovery

• metadata_objects — List all available CRM object types
• metadata_fields — Get field definitions for any object type
• metadata_picklist — Retrieve picklist values and options

🏗️ Schema Management

• object_create — Create new custom objects
• field_create — Add custom fields to existing objects

📝 Record Operations

• record_create — Create new records for any object type
• record_update — Update existing records with new values

Usage Examples

Once configured, try these natural language prompts:

Exploring Your Fireberry platform

"What object types are available in my Fireberry CRM?"
"Show me all fields for the Contacts object"
"List the picklist values for the Account Status field"

Data Operations

"Create a new custom object called 'Projects' with description, and status fields"
"Add a 'Project Budget' currency field to the Projects object"
"Create a new project record called 'Q1 Digital Transformation'"
"Import this contacts.csv file into my CRM"

Configuration

Environment Variables

Security

• 🔐 All requests authenticated with your Fireberry API token
• 🔑 Token validation on startup

Troubleshooting

Common Issues

Server not starting?

• Verify your FIREBERRY_TOKEN_ID is correct
• Check that Node.js/Bun is properly installed
• Ensure network connectivity to api.fireberry.com

Tools not appearing?

• Restart your AI assistant after configuration
• Verify JSON syntax in configuration files
• Check MCP server logs for error messages

Development Setup

git clone https://github.com/fireberry/mcp-server
cd mcp-server
npm install
npm run dev

License

MIT License - see LICENSE file for details.