arcade-mcp/libs/arcade-mcp-server/docs/advanced/transports.md
Eric Gustin 3424ec8219
MCP Local (#563)
Versions:
* arcade-mcp\==1.0.0rc1
* arcade-mcp-server\==1.0.0rc1
* arcade-core\==2.5.0rc1
* arcade-tdk\==2.6.0rc1
* arcade-serve\==2.2.0rc1

### Summary
Adds first-class MCP support across Arcade, introduces a new MCP server
and CLI, unifies the project under the arcade-mcp name, overhauls
templates/scaffolding, and improves developer tooling, secrets
management, and examples.

### Highlights
- **MCP Server & Core**
- New MCP server with stdio and HTTP/SSE transports, session management,
resumability, and lifecycle handling.
- FastAPI-like `MCPApp` for building servers with lazy init; integrated
worker+MCP HTTP app option.
- Middleware system (logging and error handling), robust exception
hierarchy, and Pydantic-based settings.
- Async-safe managers for tools, resources, and prompts backed by
registries and locks.
- Developer-facing, transport-agnostic runtime context interfaces (logs,
tools, prompts, resources, sampling, UI, notifications).
- Conversion from Arcade ToolDefinition to MCP tool schema; OpenAI JSON
tool schema converter.
  - Parser supports `@app.tool`/`@app.tool(...)` decorators.

- **CLI**
  - New `mcp` command to run MCP servers with stdio or HTTP/SSE.
- New `secret` command to set/list/unset tool secrets (supports .env
input, preserves original casing for lookups).
- `new` command refactored; option to create a full toolkit package with
scaffolding.
  - `chat` command removed.
- `serve.py` imports updated to `arcade_serve.fastapi.telemetry`;
version retrieval now uses `arcade-mcp`.
  - `show.py` refactor to use new local catalog utilities.
- `display_tool_details` improved: adds “Default” column and handles
nested properties.

- **Configuration & Discovery**
- New `configure.py` to set up Claude Desktop, Cursor, and VS Code to
connect to local or Arcade Cloud MCP servers.
- Discovery utilities to find/install toolkits, build `ToolCatalog`s,
analyze files for tools, load kits from directories (pyproject parsing),
and build minimal toolkits.
- Better handling of provider API key resolution and evaluation suite
loading.

- **Templates & Scaffolding**
- Reorganized template structure (minimal vs full); moved
`.pre-commit-config.yaml`, `.ruff.toml`, license, Makefile, README,
tests, and tools layout to correct paths.
  - Minimal template adds `.env.example` for runtime secret injection.
- Template pyproject updated for MCP servers; includes sample server
with greeting and secret-reveal tools.
  - Authorization flow in templates simplified.

- **Repo-wide Renaming & Examples**
- Migrates references from `arcade-ai` to `arcade-mcp` across READMEs,
scripts, and package metadata.
- Examples updated (LangChain/LangGraph/AI SDK/TypeScript) and package
name changed to `arcade-mcp-sdk`.

- **Evals & Core Utilities**
- Evals now use OpenAI tooling format (`OpenAIToolList`, `to_openai`);
`tool_eval` takes `provider_api_key`.
- Core utilities: fixed `does_function_return_value` by dedenting before
parse; version bump to `2.5.0rc1` and dependency cleanup.

- **Tooling & CI**
- `setup-uv-env` action splits toolkit vs contrib dependency
installation.
- Pre-commit: excludes `libs/arcade-mcp-server/mkdocs.yml` and
`libs/tests/` from YAML and Ruff hooks; Ruff per-file ignores (e.g.,
C901 in `libs/**/*.py`, TRY400 in server docs paths).
- Makefile updates for uv env setup, quality checks, tests, builds, and
new `shell` target.
  - Added Makefile to MCP server library to streamline dev workflow.

- **Cleanup**
  - Removed `claude.json` config.
- Simplified stdio entrypoint; removed unused imports (`arcade_gmail`,
`arcade_search`).

### Breaking Changes
- **CLI**: `chat` command removed; use `mcp`, `secret`, and updated
`new`.
- **Naming**: All users should update references from `arcade-ai` to
`arcade-mcp`.
- **Templates**: File paths moved; downstream scripts referencing old
template locations may need updates.

### Getting Started
- Run an MCP server:
  - `arcade mcp --stdio --toolkits your_toolkit`
  - `arcade mcp --http --toolkits your_toolkit`
- Manage secrets:
  - `arcade secret set your_toolkit KEY=value`
  - `arcade secret list your_toolkit`
  - `arcade secret unset your_toolkit KEY`
- Configure clients:
- `arcade configure` to set up Claude Desktop, Cursor, and VS Code for
local/Arcade Cloud MCP.

---------

Co-authored-by: Sam Partee <sam@arcade-ai.com>
Co-authored-by: Shub <125150494+shubcodes@users.noreply.github.com>
2025-09-25 15:28:15 -07:00

4.1 KiB

Transport Modes

MCP servers can communicate with clients through different transport mechanisms. Each transport is optimized for specific use cases and client types.

stdio Transport

The stdio (standard input/output) transport is used for direct client connections.

Characteristics

  • Communicates via standard input/output streams
  • Logs go to stderr to avoid interfering with protocol messages
  • Ideal for desktop applications and command-line tools
  • Used by Claude Desktop and similar clients

Usage

# Run with stdio transport
python -m arcade_mcp_server stdio

# Or with MCPApp
app.run(transport="stdio")

Client Configuration

For Claude Desktop, configure in ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "my-tools": {
      "command": "python",
      "args": ["-m", "arcade_mcp_server", "stdio"],
      "cwd": "/path/to/your/tools"
    }
  }
}

HTTP Transport

The HTTP transport provides REST/SSE endpoints for web-based clients.

Characteristics

  • RESTful API with Server-Sent Events (SSE) for streaming
  • Supports hot reload for development
  • Includes health checks and API documentation
  • Can be deployed behind reverse proxies
  • Suitable for web applications and services

Usage

# Run with HTTP transport (default)
python -m arcade_mcp_server

# With specific host and port
python -m arcade_mcp_server --host 0.0.0.0 --port 8080

# Or with MCPApp
app.run(transport="http", host="0.0.0.0", port=8080)

Endpoints

When running in HTTP mode, the server provides:

  • GET /health - Health check endpoint
  • GET /mcp - SSE endpoint for MCP protocol
  • GET /docs - Swagger UI documentation (debug mode)
  • GET /redoc - ReDoc documentation (debug mode)

Development Features

# Enable hot reload and debug mode
python -m arcade_mcp_server --reload --debug

# This enables:
# - Automatic restart on code changes
# - Detailed error messages
# - API documentation endpoints
# - Verbose logging

Choosing a Transport

Use stdio when:

  • Integrating with desktop applications (Claude Desktop, VS Code)
  • Building command-line tools
  • You need simple, direct communication
  • Running in environments without network access

Use HTTP when:

  • Building web applications
  • Deploying to cloud environments
  • You need to support multiple concurrent clients
  • Integrating with existing web services
  • You want API documentation and testing tools

Transport Configuration

Environment Variables

Both transports respect common environment variables:

# Server identification
MCP_SERVER_NAME="My MCP Server"
MCP_SERVER_VERSION="1.0.0"

# Logging
MCP_DEBUG=true
MCP_LOG_LEVEL=DEBUG

# HTTP-specific
MCP_HTTP_HOST=0.0.0.0
MCP_HTTP_PORT=8080

Programmatic Configuration

When using MCPApp:

from arcade_mcp_server import MCPApp

app = MCPApp(
    name="my-server",
    version="1.0.0",
    log_level="DEBUG"
)

# Run with specific transport
if __name__ == "__main__":
    import sys

    if len(sys.argv) > 1 and sys.argv[1] == "stdio":
        app.run(transport="stdio")
    else:
        app.run(transport="http", host="0.0.0.0", port=8080)

Security Considerations

stdio Transport

  • Inherits security context of the parent process
  • No network exposure
  • Suitable for trusted environments

HTTP Transport

  • Exposes network endpoints
  • Should use authentication in production
  • Consider using HTTPS with reverse proxy
  • Implement rate limiting for public deployments

Advanced Transport Features

Custom Middleware (HTTP)

Add custom middleware to HTTP transports:

from arcade_mcp_server import MCPApp

app = MCPApp(name="my-server")

# Add custom middleware
@app.middleware("http")
async def add_custom_headers(request, call_next):
    response = await call_next(request)
    response.headers["X-Custom-Header"] = "value"
    return response

Transport Events

Listen to transport lifecycle events:

@app.on_event("startup")
async def startup_handler():
    print("Server starting up...")

@app.on_event("shutdown")
async def shutdown_handler():
    print("Server shutting down...")