Skip to main content
Glama
Super Fireberry MCP

Super Fireberry MCP

by TheOctMind

Super Fireberry MCP


Overview

Super Fireberry MCP is a production-grade implementation of the Model Context Protocol (MCP) specifically engineered for Fireberry CRM. It provides a high-fidelity interface for Large Language Models (LLMs) to interact with CRM data using natural language.

This server acts as an intelligent middleware, abstracting the complexities of the Fireberry REST API into a set of powerful, self-documenting tools.

Related MCP server: Follow Up Boss MCP Server

Architecture

The project is built on a modular service-oriented architecture, ensuring high performance and type safety.

graph TD
    A[LLM / Claude] -->|MCP Protocol| B[Super Fireberry Server]
    B -->|Normalization Layer| C[API Service]
    C -->|Authenticated REST| D[Fireberry CRM API]
    D -->|JSON Response| C
    C -->|Typed Objects| B
    B -->|Structured Output| A

Key Components

  • Normalization Service: Maps human-readable entity names to internal system codes.

  • Query Engine: Translates natural language intent into structured Fireberry Query API payloads.

  • Batch Handler: Manages high-volume data operations with efficient error recovery.

Features

  • Full Metadata Reflection: Zero-config support for custom objects and fields.

  • Power Query: Advanced filtering (logical AND/OR), sorting, and deep pagination.

  • Financial Integration: Native support for Invoices, Receipts, and Transaction Items.

  • Batch Processing: High-throughput creation and updates for enterprise data loads.

  • Relationship Intelligence: Navigate complex record hierarchies effortlessly.

Toolset

Tool

Capability

Typical Use Case

get_objects

System Discovery

"What modules are available?"

get_fields

Schema Inspection

"What are the fields for 'Projects'?"

query

Advanced Search

"Find all leads from LinkedIn with status 'New'."

manage_record

Full CRUD

"Update the status of Account #105 to 'Active'."

get_related_records

Hierarchy Traversal

"Get all tasks linked to this contact."

Installation & Setup

Prerequisites

  • Node.js v18.0.0 or higher

  • A valid Fireberry API Access Token

Local Setup

git clone https://github.com/TheOctMind/super-fireberry-mcp.git
cd super-fireberry-mcp
npm install
npm run build

Integration

Claude Desktop Configuration

Add the following entry to your claude_desktop_config.json:

{
  "mcpServers": {
    "super-fireberry": {
      "command": "node",
      "args": ["/absolute/path/to/super-fireberry-mcp/build/index.js"],
      "env": {
        "FIREBERRY_TOKEN_ID": "YOUR_API_TOKEN"
      }
    }
  }
}

Roadmap

  • v1.1: Direct file attachment streaming.

  • v1.2: Webhook integration for real-time notifications.

  • v2.0: Native support for Fireberry Workflow execution.

Contributing

We welcome contributions! Please see our Contributing Guidelines for more details.

License

This project is licensed under the MIT License - see the LICENSE file for details.


Available Tools

5 tools
get_fieldsA

Retrieve detailed field definitions for any object. Essential before creating or updating records to know which fields exist.

Inputs: objectType (required).

get_objectsA

Discover all available CRM modules and custom objects. Use this first to understand the system structure.

Inputs: no named parameters.

manage_recordA

Create, Update, or Delete CRM records. For 'update', recordId is mandatory. For 'create', include the data object with field values.

Inputs: action (required), objectType (required), data, recordId.

queryC

Execute complex search queries with logical filters. Supports pageNumber, pageSize, and filter strings (e.g., 'accountname contains "test"').

Inputs: query (required), objectType (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 5 tools

    Disambiguation5/5

    Each tool addresses a clearly distinct layer: object metadata, field metadata, record search, record mutations, and relationship traversal. Even though query and get_related_records both return records, their purposes are separated by general filtered search versus direct linked-record navigation.

    Naming Consistency3/5

    The get_objects, get_fields, and get_related_records tools follow a consistent get_ prefix, but query is a bare noun/verb and manage_record uses a generic 'manage' verb. The naming is readable but mixes conventions rather than following a uniform verb_noun pattern.

    Tool Count5/5

    Five tools is a well-scoped count for a CRM MCP server. Each tool covers a major capability—discovery, metadata, search, mutations, and relationships—without unnecessary fragmentation or overwhelming breadth.

    Completeness4/5

    The tool set covers metadata discovery, record creation/updating/deletion, complex search, and relationship traversal, which covers core CRM workflows. Minor gaps exist, such as no explicit single-record fetch and no relationship mutation tools, but query and manage_record can likely work around these.

    Maintenance

    ActivityInactive
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      A production-ready MCP server for Method CRM API integration. It enables LLMs to interact with Method CRM data through tools for tables, files, users, events, and API key management.
      3
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      A production-ready MCP server that exposes the Salesmate CRM to Claude Desktop, enabling search, update, and creation of contacts, deals, tasks, notes, and activities.
      8
      1
      MIT