MCP ExplorerExplorer

Fibery MCP Server

@Fibery-incon 10 days ago
18 MIT
FreeOfficial
Productivity
#fibery#mcp#llm
This MCP (Model Context Protocol) server provides integration between Fibery and any LLM provider supporting the MCP protocol (e.g., Claude for Desktop), allowing you to interact with your Fibery workspace using natural language.

Overview

What is Fibery MCP Server

Fibery MCP Server is an integration tool that connects Fibery, a collaborative workspace platform, with various LLM providers supporting the Model Context Protocol (MCP). It allows users to interact with their Fibery workspace using natural language processing to streamline data access and manipulation.

Use cases

Users can utilize the Fibery MCP Server to query data within their Fibery entities, discover database structures, and perform CRUD operations (Create, Read, Update, Delete) interactively through a conversational interface, making it easier to manage and manipulate their data.

How to use

To use the Fibery MCP Server, users can install it via Smithery or UV. After installation, they must configure their MCP client with their Fibery domain and API token. Users can then access tools to list databases, describe database structures, query data, and create or update entities directly from their chosen LLM client.

Key features

Key features of the Fibery MCP Server include natural language queries for entities, comprehensive information retrieval about databases and fields, and the ability to create and update entities through conversational commands, enhancing user interaction with the Fibery platform.

Where to use

The Fibery MCP Server is designed for use within the Fibery workspace environment, particularly in conjunction with LLM providers like Claude for Desktop, where users can leverage natural language to perform various data management tasks seamlessly.

Content

Fibery MCP Server

smithery badge

This MCP (Model Context Protocol) server provides integration between Fibery and any LLM provider supporting the MCP protocol (e.g., Claude for Desktop), allowing you to interact with your Fibery workspace using natural language.

✨ Features

  • Query Fibery entities using natural language
  • Get information about your Fibery databases and their fields
  • Create and update Fibery entities through conversational interfaces

📦 Installation

Installing via Smithery

To install Fibery MCP Server for Claude Desktop automatically via Smithery:

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

Installing via UV

Pre-requisites:

  • A Fibery account with an API token
  • Python 3.10 or higher
  • uv

Installation Steps:

  1. Install the tool using uv:
uv tool install fibery-mcp-server
  1. Then, add this configuration to your MCP client config file. In Claude Desktop, you can access the config in 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"
      ]
    }
  }
}

Note: If “uv” command does not work, try absolute path (i.e. /Users/username/.local/bin/uv)

For Development:

{
  "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"
      ]
    }
  }
}

🚀 Available Tools

1. List Databases (list_databases)

Retrieves a list of all databases available in your Fibery workspace.

2. Describe Database (describe_database)

Provides a detailed breakdown of a specific database’s structure, showing all fields with their titles, names, and types.

3. Query Database (query_database)

Offers powerful, flexible access to your Fibery data through the Fibery API.

4. Create Entity (create_entity)

Creates new entities in your Fibery workspace with specified field values.

5. Create Entities (create_entities_batch)

Creates multiple new entities in your Fibery workspace with specified field values.

6. Update Entity (update_entity)

Updates existing entities in your Fibery workspace with new field values.

Tools

current_date
Get today's date in ISO 8601 format (YYYY-mm-dd.HH:MM:SS.000Z)
list_databases
Get list of all databases (their names) in user's Fibery workspace (schema)
describe_database
Get list of all fields (in format of 'Title [name]: type') in the selected Fibery database and for all related databases.
query_database
Run any Fibery API command. This gives tremendous flexibility, but requires a bit of experience with the low-level Fibery API. In case query succeeded, return value contains a list of records with fields you specified in select. If request failed, will return detailed error message. Examples (note, that these databases are non-existent, use databases only from user's schema!): Query: What newly created Features do we have for the past 2 months? Tool use: { "q_from": "Dev/Feature", "q_select": { "Name": ["Dev/Name"], "Public Id": ["fibery/public-id"], "Creation Date": ["fibery/creation-date"] }, "q_where": [">", ["fibery/creation-date"], "$twoMonthsAgo"], "q_order_by": {"fibery/creation-date": "q/desc"}, "q_limit": 100, "q_offset": 0, "q_params": { $twoMonthsAgo: "2025-01-16T00:00:00.000Z" } } Query: What Admin Tasks for the past week are Approval or Done? Tool use: { "q_from": "Administrative/Admin Task", "q_select": { "Name": ["Administrative/Name"], "Public Id": ["fibery/public-id"], "Creation Date": ["fibery/creation-date"], "State": ["workflow/state", "enum/name"] }, "q_where": [ "q/and", # satisfy time AND states condition [">", ["fibery/creation-date"], "$oneWeekAgo"], [ "q/or", # nested or, since entity can be in either of these states ["=", ["workflow/state", "enum/name"], "$state1"], ["=", ["workflow/state", "enum/name"], "$state2"] ] ], "q_order_by": {"fibery/creation-date": "q/desc"}, "q_limit": 100, "q_offset": 0, "q_params": { # notice that parameters used in "where" are always passed in params! $oneWeekAgo: "2025-03-07T00:00:00.000Z", $state1: "Approval", $state2: "Done" } } Query: What Admin Tasks for the past week are Approval or Done? Tool use: { "q_from": "Administrative/Admin Task", "q_select": { "State": ["workflow/state", "enum/name"], "Public Id": ["fibery/public-id"], "Creation Date": ["fibery/creation-date"], "Modification Date": ["fibery/modification-date"], "Deadline": ["Administrative/Deadline"], "Group": ["Administrative/Group", "Administrative/name"], "Name": ["Administrative/Name"], "Priority": ["Administrative/Priority_Administrative/Admin Task", "enum/name"] }, "q_where": ["!=", ["workflow/state", "workflow/Final"], "$stateType"], # Administrative/Admin Task is not "Finished" yet "q_order_by": {"fibery/creation-date": "q/desc"}, "q_limit": 100, "q_offset": 0, "q_params: { "$stateType": true } } Query: Summarize acc contacts with public id 1. Tool use: { "q_from": "Accounting/Acc Contacts", "q_select": { "Name": ["Accounting/Name"], "Public Id": ["fibery/public-id"], "Creation Date": ["fibery/creation-date"], "Description": ["Accounting/Description"] }, "q_where": ["=", ["fibery/public-id"], "$publicId"], "q_limit": 1, "q_params": { $publicId: "1", } }
create_entity
Create Fibery entity with specified fields. Examples (note, that these databases are non-existent, use databases only from user's schema!): Query: Create a feature Tool use: { "database": "Product Management/Feature", "entity": { "Product Management/Name": "New Feature", "Product Management/Description": "Description of the new feature", "workflow/state": "To Do" } } In case of successful execution, you will get a link to created entity. Make sure to give that link to the user.
update_entity
Update Fibery entity with specified fields. Examples (note, that these databases are non-existent, use databases only from user's schema!): Query: Update a feature we talked about Tool use: { "database": "Product Management/Feature", "entity": { "fibery/id": "12345678-1234-5678-1234-567812345678", "Product Management/Name": "New Feature 2", "Product Management/Description": {"append": true, "content": "Notes: some notes"}, "workflow/state": "In Progress" } } In case of successful execution, you will get a link to updated entity. Make sure to give that link to the user.

Comments