98fad93d21
# 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>
92 lines
3.5 KiB
Python
92 lines
3.5 KiB
Python
"""Tests for arcade_core.converters.utils module."""
|
|
|
|
from arcade_core.converters.utils import denormalize_tool_name, normalize_tool_name
|
|
|
|
|
|
class TestNormalizeToolName:
|
|
"""Tests for normalize_tool_name function."""
|
|
|
|
def test_simple_dot_notation(self):
|
|
"""Test converting simple dot notation to underscores."""
|
|
assert normalize_tool_name("Google.Search") == "Google_Search"
|
|
|
|
def test_multiple_dots(self):
|
|
"""Test converting multiple dots."""
|
|
assert normalize_tool_name("Namespace.Sub.Tool") == "Namespace_Sub_Tool"
|
|
|
|
def test_no_dots(self):
|
|
"""Test that names without dots are unchanged."""
|
|
assert normalize_tool_name("MyTool") == "MyTool"
|
|
|
|
def test_empty_string(self):
|
|
"""Test empty string input."""
|
|
assert normalize_tool_name("") == ""
|
|
|
|
def test_underscore_preserved(self):
|
|
"""Test that existing underscores are preserved."""
|
|
assert normalize_tool_name("My_Tool.Name") == "My_Tool_Name"
|
|
|
|
def test_single_character(self):
|
|
"""Test single character names."""
|
|
assert normalize_tool_name("A") == "A"
|
|
assert normalize_tool_name(".") == "_"
|
|
|
|
|
|
class TestDenormalizeToolName:
|
|
"""Tests for denormalize_tool_name function."""
|
|
|
|
def test_simple_underscore_notation(self):
|
|
"""Test converting simple underscore notation to dots."""
|
|
assert denormalize_tool_name("Google_Search") == "Google.Search"
|
|
|
|
def test_multiple_underscores(self):
|
|
"""Test converting multiple underscores."""
|
|
assert denormalize_tool_name("Namespace_Sub_Tool") == "Namespace.Sub.Tool"
|
|
|
|
def test_no_underscores(self):
|
|
"""Test that names without underscores are unchanged."""
|
|
assert denormalize_tool_name("MyTool") == "MyTool"
|
|
|
|
def test_empty_string(self):
|
|
"""Test empty string input."""
|
|
assert denormalize_tool_name("") == ""
|
|
|
|
def test_custom_separator(self):
|
|
"""Test using a custom separator."""
|
|
assert denormalize_tool_name("Google_Search", separator="::") == "Google::Search"
|
|
|
|
def test_single_character(self):
|
|
"""Test single character names."""
|
|
assert denormalize_tool_name("A") == "A"
|
|
assert denormalize_tool_name("_") == "."
|
|
|
|
|
|
class TestRoundTrip:
|
|
"""Tests for round-trip conversion (normalize then denormalize)."""
|
|
|
|
def test_roundtrip_simple(self):
|
|
"""Test round-trip for simple names without original underscores."""
|
|
original = "Google.Search"
|
|
normalized = normalize_tool_name(original)
|
|
denormalized = denormalize_tool_name(normalized)
|
|
assert denormalized == original
|
|
|
|
def test_roundtrip_multiple_dots(self):
|
|
"""Test round-trip for names with multiple dots."""
|
|
original = "Namespace.Sub.Tool"
|
|
normalized = normalize_tool_name(original)
|
|
denormalized = denormalize_tool_name(normalized)
|
|
assert denormalized == original
|
|
|
|
def test_roundtrip_with_original_underscores_is_lossy(self):
|
|
"""Test that round-trip is lossy when original has underscores.
|
|
|
|
This documents the known limitation: if the original name contains
|
|
underscores, denormalization cannot distinguish them from dots.
|
|
"""
|
|
original = "My_Tool.Name"
|
|
normalized = normalize_tool_name(original) # "My_Tool_Name"
|
|
denormalized = denormalize_tool_name(normalized) # "My.Tool.Name"
|
|
# This is NOT equal to original - expected behavior
|
|
assert denormalized != original
|
|
assert denormalized == "My.Tool.Name"
|