mstream-mcp-server

An MCP server for https://github.com/makarski/mstream

Prerequisites

• Python 3.11 or later
• Recommended: a virtual environment (python -m venv .venv)

Available Tools

This MCP server provides the following tools for managing mstream jobs and services:

Job Management

• list_jobs - List all configured mstream jobs with their current status, metadata, and configuration
• create_job - Create a new mstream job with input/output schemas and batch configuration
• stop_job - Stop a running mstream job by job ID
• restart_job - Restart a stopped mstream job by job ID

Service Management

• list_services - List all registered mstream services
• get_service - Get details of a specific service by service ID
• create_service - Register a new service with mstream, including endpoint and schema definitions
• delete_service - Remove a service from mstream by service ID

Usage Examples

# List all jobs
list_jobs()

# Create a new job with schema
create_job({
  "name": "data-processing-job",
  "input_schema": {
    "name": "input",
    "fields": [
      {"name": "data", "type": "string", "required": true}
    ]
  },
  "batch_config": {
    "batch_size": 100,
    "max_concurrency": 5
  }
})

# Stop a job
stop_job("job-123")

# Create a service
create_service({
  "name": "my-service",
  "endpoint": "http://localhost:8080/process",
  "schemas": [{
    "name": "request",
    "fields": [{"name": "input", "type": "string", "required": true}]
  }]
})

Setup

1. Clone the repository and create/activate your virtual environment.

   python -m venv .venv
   . ./.venv/bin/activate
   

2. Install the package along with development dependencies:

   pip install -e ".[dev]"
   

Development

• Format and lint the codebase:

  ruff check .
  black .
  

• Type checking:

  mypy .
  

• Run tests:

  pytest
  

Running the server

The HTTP transport entrypoint is accessible via python -m mstream_mcp_server. Example:

python -m mstream_mcp_server --host 0.0.0.0 --port 8000 --api-base-url http://localhost --api-port 8700

Configuration may also be provided via environment variables:

• MSTREAM_SERVER_HOST / MSTREAM_SERVER_PORT: bind address for the MCP HTTP server.
• MSTREAM_API_BASE_URL / MSTREAM_API_PORT: address of the upstream mstream API.
• MSTREAM_API_TOKEN: bearer token for API authentication.
• MSTREAM_API_TIMEOUT, MSTREAM_API_MAX_RETRIES, MSTREAM_API_BACKOFF_FACTOR: HTTP client tuning.
• MSTREAM_LOG_LEVEL: logging verbosity for the server and transport.

VS Code Configuration

Add the following to your VS Code settings.json to connect to the mstream MCP server:

{
  "mcp.servers": {
    "mstream": {
      "type": "http",
      "url": "http://localhost:<PORT>/mcp",
      "transport": {
        "type": "http"
      }
    }
  }
}