Get up and running with MCP tools and Meshagent in a cloud Room.

​
1. Install Meshagent CLI

pip install "meshagent[all]"

​
2. Sign up and authenticate

Follow the setup guide at Meshagent CLI Getting Started.

​
3. Start a Test MCP Server Agent in a Meshagent Room

meshagent service test \
  --room=test \
  --role=agent \
  --image=meshagent/mcp-fetch:latest \
  --env MESHAGENT_PORT=8001 \
  --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
  --name=mcp-fetch-service-test
  • This command launches a Meshagent Room named test with the MCP fetch server running as an agent in the room.
  • Rooms will automatically close if they go inactive.

​
4. Start a Chatbot That Can Use the MCP Fetch Toolkit

meshagent chatbot join \
  --room=test \
  --agent-name=mcp-fetch \
  --name=mcp-fetch \
  --toolkit=mcp-fetch
  • This launches a chatbot in the same room, configured to use the fetch toolkit.
  • More than one toolkit can be enabled in the same room with the same agent.
  • A browser-accessible link to the Room is output by this command.

​
5. Interact in the Room

Open the Room link in your browser (provided in the output above), and message the agent to interact directly with the MCP Server tools.

​
Project Level Deployment

Create persistent services at the project level so MCP tools and chatbots are available in all Rooms automatically—ideal for production! No need to manually run MCP servers or chatbot processes locally every time.

​
1. Deploy the MCP Server as a Persistent Cloud Agent

meshagent service create \
  --role=agent \
  --image=meshagent/mcp-fetch:latest \
  --env MESHAGENT_PORT=8001 \
  --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
  --name=mcp-fetch-service

​
2. Deploy the Chatbot as a Persistent Cloud Service

meshagent service create \
  --image="meshagent/cli:latest" \
  --port="num=9001 path=/agent liveness=/ type=meshagent.callable participant_name=mcp-fetch-chatbot" \
  --env="MESHAGENT_PORT=9001" \
  --name="mcp-fetch-chatbot-service" \
  --command="meshagent chatbot service --agent-name=mcp-fetch-chatbot --toolkit=mcp-fetch"
  • These services will automatically join every new Room in your Meshagent project.

​
Tools Available

​
MCP Fetch (Reference) Server

Fetches a URL from the internet and extracts its contents as markdown. For more details on the MCP protocol, see What is an MCP Server?

​
Provided Tool

ToolDescription
fetchFetches a URL from the internet and optionally extracts its contents as markdown.

​
Tool Details

​
Tool: fetch

Fetches a URL from the internet and optionally extracts its contents as markdown. You now have internet access via this tool. Fetch the latest information from the web.
ParameterTypeDescription
urlstringURL to fetch
max_lengthinteger optionalMaximum number of characters to return
rawboolean optionalGet the actual HTML content without simplification
start_indexinteger optionalStart output at this index if a previous fetch was truncated