This lets you launch the Fibery MCP server and chatbot toolkit in an isolated, temporary Room for testing.

Environment Variables Required

  • FIBERY_HOST: Your Fibery workspace domain (e.g. your-domain.fibery.io)
  • FIBERY_API_TOKEN: API token for your Fibery workspace

Get Started in 5 Steps

Replace environment variable values (<your-domain>, <your-api-token>) as needed.
  1. Install Meshagent CLI:
    pip install "meshagent[all]"
    
  2. Sign up and authenticate:
    Follow this guide.
  3. Start the MCP Server in a Room:
    meshagent service test --room=test --role=agent \
      --image=meshagent/mcp-fibery:latest \
      --env MESHAGENT_PORT=8001 \
      --env FIBERY_HOST=<your-domain.fibery.io> \
      --env FIBERY_API_TOKEN=<your-api-token> \
      --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
      --name=mcp-fibery-service-test
    
    • Starts a Meshagent Room (test) with the Fibery MCP server ready. Rooms close after inactivity, cleaning up all tools and chatbots.
  4. Join a Chatbot with the Fibery MCP Toolkit:
    meshagent chatbot join --room=test --agent-name=mcp-fibery --name=mcp-fibery --toolkit=mcp-fibery
    
    • Adds an AI chatbot to the Room, equipped to use the Fibery MCP tools.
    • You can use multiple toolkits in a single Room.
    • The CLI will print a link to the Room.
  5. Test the Agent in Your Browser:
    • Visit the link produced above.
    • Interact with the agent via chat—try out the Fibery MCP tools instantly!

🏗 Project-Level Deployment (Persistent Agents & Chatbots)

For production, make the MCP server and chatbot available in all Project Rooms automatically. These services persist—no need to start agents or chatbots for each Room manually.
  1. Deploy Persistent MCP Server Service:
    meshagent service create --role=agent \
      --image=meshagent/mcp-fibery:latest \
      --env MESHAGENT_PORT=8001 \
      --env FIBERY_HOST=<your-domain.fibery.io> \
      --env FIBERY_API_TOKEN=<your-api-token> \
      --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
      --name=mcp-fibery-service
    
  2. Deploy Persistent Chatbot Service:
    meshagent service create \
      --image="meshagent/cli:latest" \
      --port="num=9001 path=/agent liveness=/ type=meshagent.callable participant_name=mcp-fibery-chatbot" \
      --env="MESHAGENT_PORT=9001" \
      --name="mcp-fibery-chatbot-service" \
      --command="meshagent chatbot service --agent-name=mcp-fibery-chatbot --toolkit=mcp-fibery"
    
  • Both the MCP server and the chatbot automatically join every new Project Room, ready for users to interact with Fibery tools.

🧰 Tools Available

Interact with your Fibery workspace directly from Meshagent Rooms!

Tools Summary

ToolShort Description
create_entities_batchCreate multiple Fibery entities at once with specified fields.
create_entityCreate Fibery entity with specified fields.
current_dateGet today’s date in ISO 8601 format (YYYY-mm-dd.HH:MM:SS.000Z)
describe_databaseGet list of all fields in the selected Fibery database and related databases.
list_databasesGet list of all databases in user’s Fibery workspace (schema)
query_databaseRun any Fibery API command. Powerful, flexible querying.
update_entityUpdate a Fibery entity with specified fields.

Tools Details

create_entities_batch

Create multiple entities in a chosen Fibery database with specified fields.
  • Parameters:
    • database (string): Database to create entities in
    • entities (object): List of dictionaries with fields to set for each entity
  • Returns: Links to created entities
  • Example:
    {
        "database": "Product Management/Feature",
        "entities": [
            {"Product Management/Name": "New Feature 1", ...},
            {"Product Management/Name": "New Feature 2", ...}
        ]
    }
    

create_entity

Create a single Fibery entity in the specified database.
  • Parameters:
    • database (string)
    • entity (object): Dictionary of fields/values
  • Returns: Link to the created entity
  • Example:
    {
        "database": "Product Management/Feature",
        "entity": {"Product Management/Name": "New Feature", ...}
    }
    

current_date

Get the current UTC date/time in ISO 8601 format.

describe_database

Returns details of all fields in a specified Fibery database (and related schemas).
  • Parameters:
    • database_name (string)

list_databases

Lists all available databases in your Fibery workspace (schema).

query_database

Run any complex or custom Fibery API query.
  • Parameters:
    • q_from (string): Entity type
    • q_select (object): Fields to retrieve
    • q_where (object, optional): Filter conditions (supports logical and comparative operators)
    • q_limit, q_offset, q_order_by, q_params (optional, for advanced usage)
  • Example:
    {
        "q_from": "Dev/Feature",
        "q_select": {"Name": ["Dev/Name"]},
        "q_where": [">", ["fibery/creation-date"], "$twoMonthsAgo"],
        "q_params": {"$twoMonthsAgo": "2025-01-16T00:00:00.000Z"}
    }
    

update_entity

Update an existing entity with new values or appended information.
  • Parameters:
    • database (string)
    • entity (object): Must include entity fibery/id; document fields can specify {append: true, content: "..."} structure for appending
  • Returns: Link to updated entity

For more information on the MCP protocol and best practices:
Happy collaborating in the cloud! 🚀