# MCP Server Tool Evaluation Support
## Overview
Add support for evaluating tools from remote MCP servers without
requiring Python callables. Enables direct evaluation of any
MCP-compatible tool server.
## What's New
### Core Features
- **`MCPToolRegistry`**: Evaluate tools from a single MCP server
- **`CompositeMCPRegistry`**: Evaluate tools from multiple MCP servers
simultaneously
- **Automatic loaders**: `load_from_stdio()` and `load_from_http()` to
fetch tools from running servers
- **Automatic namespacing**: Tools prefixed with server name (e.g.,
`server_tool_name`)
- **Smart name resolution**: Use short names if unique, full names if
ambiguous
- **OpenAI strict mode**: Automatic schema conversion prevents parameter
hallucinations
### Usage
**Automatic Loading:**
```python
from arcade_evals import load_from_stdio, MCPToolRegistry
# Load tools automatically from MCP server
tools = load_from_stdio(["npx", "-y", "@modelcontextprotocol/server-github"])
registry = MCPToolRegistry(tools)
```
**Single MCP Server:**
```python
from arcade_evals import MCPToolRegistry, ExpectedToolCall
registry = MCPToolRegistry(mcp_tools)
suite = EvalSuite(catalog=registry)
suite.add_case(
expected_tool_calls=[
ExpectedToolCall(tool_name="tool_name", args={...})
]
)
```
**Multiple MCP Servers:**
```python
from arcade_evals import CompositeMCPRegistry, load_from_stdio
# Load from multiple servers
github_tools = load_from_stdio(["npx", "-y", "@modelcontextprotocol/server-github"])
slack_tools = load_from_stdio(["npx", "-y", "@modelcontextprotocol/server-slack"])
composite = CompositeMCPRegistry(
tool_lists={
"github": github_tools,
"slack": slack_tools,
}
)
suite = EvalSuite(catalog=composite)
suite.add_case(
expected_tool_calls=[
ExpectedToolCall(tool_name="github_list_issues", args={...})
]
)
```
## Implementation
### Files Changed
- **`libs/arcade-evals/arcade_evals/registry.py`** (NEW): Registry
abstractions and implementations
- **`libs/arcade-evals/arcade_evals/loaders.py`** (NEW): Automatic tool
loading from MCP servers
- **`libs/arcade-evals/arcade_evals/eval.py`** (MODIFIED): Enhanced
`ExpectedToolCall` and evaluation logic
- **`libs/arcade-evals/arcade_evals/__init__.py`** (MODIFIED): Exported
new registries and loaders
### Key Technical Details
- Added `BaseToolRegistry` interface for abstraction
- `MCPToolRegistry` handles single server tools
- `CompositeMCPRegistry` manages multiple servers with collision
detection
- `load_from_stdio()` and `load_from_http()` for automatic tool
discovery
- Fixed name normalization bug: MCP tools use underscores (not dots)
- Optimized tool copying: 2.5x faster via shallow copy
## Testing
- ✅ 41 tests passing (25 new tests added)
- ✅ `test_eval_mcp_registry.py`: MCPToolRegistry functionality
- ✅ `test_eval_composite_mcp.py`: CompositeMCPRegistry with multiple
servers
- ✅ Verified backward compatibility with Python tools
## Backward Compatibility
✅ **100% backward compatible** - No breaking changes
## Breaking Changes
**None**
<!-- CURSOR_SUMMARY -->
---
> [!NOTE]
> Adds end-to-end eval UX: examples, a robust CLI runner, and rich
outputs.
>
> - **New examples**: `eval_arcade_gateway.py`,
`eval_stdio_mcp_server.py`, `eval_http_mcp_server.py`,
`eval_comprehensive_comparison.py` with timeouts, error handling, and
track-based comparisons; detailed `README.md`
> - **CLI runner**: `arcade_cli/evals_runner.py` to execute
evals/capture in parallel with progress, error isolation, failed-only
filtering, context inclusion, and multi-provider/model support
> - **Output formatters**: `arcade_cli/formatters/` (txt, md, html,
json) for evals and capture; comparative and multi-model HTML with tabs
and context rendering
> - **Display refactor**: `display.py` now supports writing multiple
formats, failed-only disclaimers, include-context, and improved console
summaries
>
> <sup>Written by [Cursor
Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit
ff8acf9c34a6b61462a019a1ee9df081006517d0. This will update automatically
on new commits. Configure
[here](https://cursor.com/dashboard?tab=bugbot).</sup>
<!-- /CURSOR_SUMMARY -->
---------
Co-authored-by: Francisco Liberal <francisco@arcade.dev>
Co-authored-by: Mateo Torres <torresmateo@gmail.com>
78 lines
2.7 KiB
Python
78 lines
2.7 KiB
Python
"""Test that MCP routes appear in OpenAPI documentation."""
|
|
|
|
from arcade_core import ToolCatalog
|
|
from arcade_core.toolkit import Toolkit
|
|
from arcade_mcp_server.settings import MCPSettings
|
|
from arcade_mcp_server.worker import create_arcade_mcp
|
|
from fastapi.testclient import TestClient
|
|
|
|
|
|
def test_mcp_routes_in_openapi(monkeypatch):
|
|
"""Test that MCP routes appear in FastAPI OpenAPI documentation."""
|
|
# Set environment variables for settings
|
|
monkeypatch.setenv("ARCADE_AUTH_DISABLED", "true")
|
|
monkeypatch.setenv("ARCADE_WORKER_SECRET", "test")
|
|
monkeypatch.setenv("MCP_SERVER_NAME", "test-mcp")
|
|
monkeypatch.setenv("MCP_SERVER_VERSION", "0.1.0")
|
|
|
|
# Create a simple catalog
|
|
catalog = ToolCatalog()
|
|
toolkit = Toolkit(name="test", package_name="test", version="0.1.0", description="Test toolkit")
|
|
catalog.add_toolkit(toolkit)
|
|
|
|
# Create MCP settings from environment
|
|
mcp_settings = MCPSettings.from_env()
|
|
|
|
# Create the app
|
|
app = create_arcade_mcp(catalog, mcp_settings=mcp_settings)
|
|
|
|
# Create test client
|
|
client = TestClient(app)
|
|
|
|
# Get OpenAPI schema
|
|
response = client.get("/openapi.json")
|
|
assert response.status_code == 200
|
|
|
|
openapi_schema = response.json()
|
|
|
|
# Check that MCP paths are documented
|
|
assert "/mcp/" in openapi_schema["paths"]
|
|
|
|
mcp_path = openapi_schema["paths"]["/mcp/"]
|
|
|
|
# Check POST endpoint
|
|
assert "post" in mcp_path
|
|
assert mcp_path["post"]["summary"] == "Send MCP message"
|
|
assert "MCPRequest" in str(mcp_path["post"])
|
|
assert "MCPResponse" in str(mcp_path["post"])
|
|
|
|
# Check GET endpoint
|
|
assert "get" in mcp_path
|
|
assert mcp_path["get"]["summary"] == "Establish SSE stream"
|
|
|
|
# Check DELETE endpoint
|
|
assert "delete" in mcp_path
|
|
assert mcp_path["delete"]["summary"] == "Terminate session"
|
|
|
|
# Check that component schemas are defined
|
|
components = openapi_schema.get("components", {}).get("schemas", {})
|
|
assert "MCPRequest" in components
|
|
assert "MCPResponse" in components
|
|
|
|
# Verify MCPRequest schema
|
|
mcp_request = components["MCPRequest"]
|
|
assert "jsonrpc" in mcp_request["properties"]
|
|
assert "method" in mcp_request["properties"]
|
|
assert "params" in mcp_request["properties"]
|
|
assert "id" in mcp_request["properties"]
|
|
|
|
# Verify that the paths include the MCP tag
|
|
assert "tags" in mcp_path["post"]
|
|
assert "MCP Protocol" in mcp_path["post"]["tags"]
|
|
|
|
# Verify the actual proxy is mounted (not routes)
|
|
# The OpenAPI docs should exist but not interfere with the mount
|
|
|
|
mounts = [route for route in app.routes if hasattr(route, "app") and hasattr(route, "path")]
|
|
mcp_mounts = [m for m in mounts if m.path == "/mcp"]
|
|
assert len(mcp_mounts) == 1, "Should have exactly one mount at /mcp"
|