Fibery MCP Gateway Service

This specialized MCP (Model Context Protocol) intermediary establishes a connection between conversational AI agents (like Claude) and your personal or team Fibery environment. It translates plain English instructions into precise API calls to manage your Fibery knowledge base.

🌟 Core Capabilities

• Semantic Data Retrieval: Ask questions about your Fibery entities in conversational terms.
• Schema Exploration: Obtain detailed metadata regarding available databases and their field definitions.
• Data Manipulation: Programmatically add new items or adjust existing records via dialogue.

📦 Deployment Instructions

Automated Setup via Smithery

For seamless, automated deployment tailored for clients like Claude Desktop, utilize the Smithery CLI:

bash npx -y @smithery/cli install @Fibery-inc/fibery-mcp-server --client claude

Manual Setup Utilizing 'uv'

Prerequisites:

1. Access credentials for the Fibery API (an active token).
2. A functional installation of Python version 3.10 or newer.
3. The package manager utility uv installed.

Installation Sequence:

1. Execute the installation command for the server tool: bash uv tool install fibery-mcp-server

2. Integrate the following configuration block into your MCP client's configuration file (accessible in Claude Desktop via Settings → Developer → Edit Config):

{ "mcpServers": { "fibery-mcp-server": { "command": "uv", "args": [ "tool", "run", "fibery-mcp-server", "--fibery-host", "your-domain.fibery.io", "--fibery-api-token", "your-api-token" ] } } }

Troubleshooting Tip: If the uv executable path is not in your system's PATH environment variable, you may need to supply the absolute location (e.g., /Users/username/.local/bin/uv).

For Development/Local Testing:

{ "mcpServers": { "fibery-mcp-server": { "command": "uv", "args": [ "--directory", "path/to/cloned/fibery-mcp-server", "run", "fibery-mcp-server", "--fibery-host", "your-domain.fibery.io", "--fibery-api-token", "your-api-token" ] } } }

🛠️ Exposed Operations

1. Retrieve Available Datasets (list_databases)

Fetches and returns a catalog of all accessible data schemas (databases) within the connected Fibery workspace.

2. Inspect Schema Details (describe_database)

Generates comprehensive structural information for a designated database, detailing every associated attribute, including its descriptive label, unique identifier, and data type.

3. Execute Data Fetch (query_database)

Provides an advanced mechanism for retrieving data from Fibery structures, leveraging the underlying Fibery API based on natural language prompts.

4. Record Addition (create_entity)

Allows for the instantiation of a single new record within a specified Fibery collection, populated according to provided attribute values.

5. Bulk Record Insertion (create_entities_batch)

Efficiently handles the creation of numerous new items within a Fibery collection simultaneously, using specified field mappings.

6. Record Modification (update_entity)

Modifies the contents of existing records in the Fibery repository by applying updated values to specified attributes.