Follow these steps to run an MCP agent and its toolkit in a Meshagent Room in the cloud for quick testing or demos:
  1. Install Meshagent CLI
    pip install "meshagent[all]"
    
  2. Sign up and authenticate
  3. Start the MCP SQLite server as a room-level agent
    meshagent service test --room=test --role=agent --image=meshagent/mcp-sqlite:latest --env MESHAGENT_PORT=8001 --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" --name=mcp-sqlite-service-test
    
    This command creates a Meshagent Room named test and deploys a test MCP SQLite server agent. Rooms close automatically when inactive.
  4. Start a chatbot in the room with the SQLite toolkit
    meshagent chatbot join --room=test --agent-name=mcp-sqlite --name=mcp-sqlite --toolkit=mcp-sqlite
    
    • Multiple toolkits can be enabled in a single room for the same agent.
    • The output of this command will include a direct link to the Meshagent Room.
  5. Test the integration
    • Open the provided room link in your browser and send a message to the agent to interact with the MCP SQLite server tools.

Project-Level Deployment

For persistent, production-ready deployment, register services at the Meshagent project level. This ensures your MCP server tools and chatbot are automatically available in every new room created within the project—no need to host services or bots locally.
  1. Deploy the MCP SQLite server at the project level
    meshagent service create --role=agent --image=meshagent/mcp-sqlite:latest --env MESHAGENT_PORT=8001 --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" --name=mcp-sqlite-service
    
  2. Deploy the chatbot at the project level
    meshagent service create --image="meshagent/cli:latest" --port="num=9001 path=/agent liveness=/ type=meshagent.callable participant_name=mcp-sqlite-chatbot" --env="MESHAGENT_PORT=9001" --name="mcp-sqlite-chatbot-service" --command="meshagent chatbot service --agent-name=mcp-sqlite-chatbot --toolkit=mcp-sqlite"
    
When any room is created in your Meshagent project, both the MCP SQLite server and chatbot will auto-join and be instantly available to participants.

Tools Available

The following tools are exposed by the MCP SQLite Server as callable actions within Meshagent Rooms:
ToolShort Description
append_insightAdd a business insight to the memo
create_tableCreate a new table in the SQLite database
describe_tableGet the schema information for a specific table
list_tablesList all tables in the SQLite database
read_queryExecute a SELECT query on the SQLite database
write_queryExecute an INSERT, UPDATE, or DELETE query

Tools Details

append_insight

Add a business insight to the memo
  • Parameters:
    • insight (string): Business insight discovered from data analysis

create_table

Create a new table in the SQLite database
  • Parameters:
    • query (string): CREATE TABLE SQL statement

describe_table

Get the schema information for a specific table
  • Parameters:
    • table_name (string): Name of the table to describe

list_tables

List all tables in the SQLite database
  • No parameters

read_query

Execute a SELECT query on the SQLite database
  • Parameters:
    • query (string): SELECT SQL query to execute

write_query

Execute an INSERT, UPDATE, or DELETE query on the SQLite database
  • Parameters:
    • query (string): SQL query to execute

Additional Resources

For questions, open an issue or reach out via the listed resources.