Follow these steps to quickly spin up a cloud-connected Meshagent Room with the Sentry MCP server and chatbot fully integrated for testing:

Prerequisites

Required Environment Variables

  • SENTRY_AUTH_TOKEN: Your Sentry API token

1. Install Meshagent with Room and tool support

pip install "meshagent[all]"

2. Sign up and authenticate with Meshagent

See: Meshagent CLI Getting Started
meshagent login

3. Launch the Sentry MCP Agent in a Room

Replace the token in <YOUR_SENTRY_TOKEN> as required.
meshagent service test \
    --room=test \
    --role=agent \
    --image=meshagent/mcp-sentry:latest \
    --env SENTRY_AUTH_TOKEN=<YOUR_SENTRY_TOKEN> \
    --env MESHAGENT_PORT=8001 \
    --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
    --name=mcp-sentry-service-test
  • What this does:
    Launches a Meshagent Room named test, running the Sentry MCP agent as a callable tool.
  • Lifecycle note:
    The MCP Sentry server and chatbot are cleaned up automatically when the Room is closed or goes inactive.

4. Start a Chatbot Connected to the Room

meshagent chatbot join \
    --room=test \
    --agent-name=mcp-sentry \
    --name=mcp-sentry \
    --toolkit=mcp-sentry
  • What this does:
    Joins the Room with a chatbot that can use the Sentry MCP toolkit.
  • Multiple Toolkits:
    You can use multiple toolkits in the same Room with the same agent.
  • Output:
    A link to the Room will be provided for browser access.

5. Try It!

  • Visit the Room link from the command output.
  • Start a chat with the agent; use plain English requests (e.g., “Retrieve details for Sentry issue 12345”).
  • The chatbot will invoke the MCP Sentry tools running in the Room, retrieving live results from your Sentry account.

Project Level Deployment (Persistent, Auto-Joint Services)

For production or always-on availability, you can deploy services at the Meshagent project level. Whenever a Room is created in your Meshagent project, both the Sentry MCP server and its chatbot are instantly available—no manual setup needed and no need to run components locally.

1. Create a Persistent Sentry MCP Server Service

meshagent service create \
    --role=agent \
    --image=meshagent/mcp-sentry:latest \
    --env SENTRY_AUTH_TOKEN=<YOUR_SENTRY_TOKEN> \
    --env MESHAGENT_PORT=8001 \
    --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
    --name=mcp-sentry-service
  • This service is now available in all Rooms in your Meshagent project.

2. Create a Persistent Sentry Chatbot Service

meshagent service create \
    --image="meshagent/cli:latest" \
    --port="num=9001 path=/agent liveness=/ type=meshagent.callable participant_name=mcp-sentry-chatbot" \
    --env="MESHAGENT_PORT=9001" \
    --name="mcp-sentry-chatbot-service" \
    --command="meshagent chatbot service --agent-name=mcp-sentry-chatbot --toolkit=mcp-sentry"
  • Every new Room in your project will auto-join this chatbot, capable of using the Sentry MCP tools.
  • You no longer need to run the MCP server or chatbot locally, or set them up each time a Room is launched.

Tools Available

Summary

ToolShort Description
get_sentry_issueRetrieve and analyze a Sentry issue by ID or URL.

Tool Details

Tool: get_sentry_issue

Retrieve and analyze a Sentry issue by ID or URL. Use this tool when you need to:
  • Investigate production errors and crashes
  • Access detailed stacktraces from Sentry
  • Analyze error patterns and frequencies
  • Get information about when issues first/last occurred
  • Review error counts and status
ParameterTypeDescription
issue_id_or_urlstringSentry issue ID or URL to analyze

Learn More