fleet-memory/hindsight-api/tests/test_mcp_extension.py
DK09876 e798979733
Harden MCP server: fix routing, validation, and usage metering (#341)
* fix: move mental model usage metering into engine for MCP support

Mental model validation hooks (validate_mental_model_get, validate_mental_model_refresh)
were only called in REST HTTP handlers, not in the engine. MCP tools call engine methods
directly, so usage metering was skipped entirely for MCP mental model operations.

Moved pre-validation and post-completion hooks into memory_engine.py (matching the
retain/recall/reflect pattern) and removed the duplicate code from http.py.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: remove double validation from create_mental_model and add internal checks

- Remove pre-validation from create_mental_model since callers always call
  submit_async_refresh_mental_model next (which validates), preventing
  double credit checks
- Add is_internal checks to mental model metering validators (matching
  the existing pattern for recall/reflect) so background worker tasks
  skip billing

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: prevent 307 redirect on /mcp that breaks MCP tool discovery

Starlette's Mount class redirects /mcp to /mcp/ with a 307 Temporary
Redirect. Many MCP clients don't follow POST redirects, which causes
tool discovery to fail (0 tools discovered despite successful auth).

Add _MCPPathRewriteMiddleware that rewrites /mcp to /mcp/ at the ASGI
level before routing, preventing the redirect entirely. Both /mcp and
/mcp/ now work identically.

Add regression test test_mcp_no_trailing_slash_works to verify URLs
with and without trailing slashes discover tools correctly.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* harden MCP server for real-world usage

- Remove MCP_ENDPOINTS blocklist so banks named "sse"/"messages" route correctly
- Scope SSE body rewriting to text/event-stream responses only to prevent data corruption
- Add _validate_mental_model_inputs for name, source_query, max_tokens validation in MCP tools
- Improve "not found" error messages to include bank_id context
- Fix fragile tool count assertions (exact → minimum bounds)
- Add integration tests: tool execution, input validation, edge-case bank names
- Add unit tests for validation helper and tool-level validation

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: replace Mount + rewrite middleware with wrapping middleware

Starlette's Mount class redirects /mcp -> /mcp/ with 307, which MCP clients
don't follow. Previously we patched this with _MCPPathRewriteMiddleware.

Now MCPMiddleware wraps the FastAPI app directly via add_middleware, intercepting
/mcp* requests before they reach Starlette's router. No Mount means no redirect.

- Remove _MCPPathRewriteMiddleware (no longer needed)
- Remove app.mount() call
- Add prefix parameter to MCPMiddleware
- Use app.add_middleware() for proper Starlette integration
- Simplify path stripping (just remove prefix, no mount/root_path handling)
- Update routing test to match current behavior (no MCP_ENDPOINTS blocklist)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: update stale docstring referencing removed _MCPPathRewriteMiddleware

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 10:41:20 +01:00

169 lines
6.1 KiB
Python

"""Tests for MCPExtension loading and tool registration."""
from unittest.mock import MagicMock, patch
import pytest
from fastmcp import FastMCP
from hindsight_api import MemoryEngine
from hindsight_api.extensions.mcp import MCPExtension
class MockMCPExtension(MCPExtension):
"""Test extension that registers a custom tool."""
def __init__(self, config=None):
super().__init__(config)
self.register_tools_called = False
self.registered_mcp = None
self.registered_memory = None
def register_tools(self, mcp: FastMCP, memory: MemoryEngine) -> None:
"""Register a test tool to verify extension was called."""
self.register_tools_called = True
self.registered_mcp = mcp
self.registered_memory = memory
@mcp.tool()
async def test_extension_tool(query: str) -> str:
"""A test tool registered by the extension."""
return f"Extension tool received: {query}"
class TestMCPExtensionBase:
"""Tests for MCPExtension base class."""
def test_mcp_extension_is_abstract(self):
"""MCPExtension.register_tools is abstract and must be implemented."""
with pytest.raises(TypeError, match="abstract method"):
MCPExtension()
def test_subclass_can_be_instantiated(self):
"""Subclass implementing register_tools can be instantiated."""
ext = MockMCPExtension()
assert ext is not None
assert ext.register_tools_called is False
def test_register_tools_receives_mcp_and_memory(self):
"""register_tools receives FastMCP and MemoryEngine instances."""
ext = MockMCPExtension()
mcp = FastMCP("test")
memory = MagicMock(spec=MemoryEngine)
ext.register_tools(mcp, memory)
assert ext.register_tools_called is True
assert ext.registered_mcp is mcp
assert ext.registered_memory is memory
class TestMCPExtensionLoading:
"""Tests for MCPExtension loading in create_mcp_server."""
@pytest.fixture
def mock_memory(self):
"""Create a mock MemoryEngine."""
memory = MagicMock()
memory._tenant_extension = MagicMock()
memory._tenant_extension.authenticate_mcp = MagicMock()
return memory
def test_create_mcp_server_without_extension(self, mock_memory):
"""create_mcp_server works without MCPExtension configured."""
from hindsight_api.api.mcp import create_mcp_server
with patch("hindsight_api.api.mcp.load_extension", return_value=None):
mcp = create_mcp_server(mock_memory)
# Core tools should be registered
tools = mcp._tool_manager._tools
assert "retain" in tools
assert "recall" in tools
assert "reflect" in tools
# Extension tool should NOT be present
assert "test_extension_tool" not in tools
def test_create_mcp_server_with_extension(self, mock_memory):
"""create_mcp_server loads and calls MCPExtension when configured."""
from hindsight_api.api.mcp import create_mcp_server
mock_ext = MockMCPExtension()
with patch("hindsight_api.api.mcp.load_extension", return_value=mock_ext):
mcp = create_mcp_server(mock_memory)
# Extension should have been called
assert mock_ext.register_tools_called is True
# Core tools should still be registered
tools = mcp._tool_manager._tools
assert "retain" in tools
assert "recall" in tools
# Extension tool should also be registered
assert "test_extension_tool" in tools
@pytest.mark.asyncio
async def test_extension_tool_is_callable(self, mock_memory):
"""Tool registered by extension can be called."""
from hindsight_api.api.mcp import create_mcp_server
mock_ext = MockMCPExtension()
with patch("hindsight_api.api.mcp.load_extension", return_value=mock_ext):
mcp = create_mcp_server(mock_memory)
# Get and call the extension tool
tools = mcp._tool_manager._tools
test_tool = tools["test_extension_tool"]
result = await test_tool.fn(query="hello world")
assert result == "Extension tool received: hello world"
def test_load_extension_called_with_correct_args(self, mock_memory):
"""load_extension is called with 'MCP' prefix and MCPExtension class."""
from hindsight_api.api.mcp import create_mcp_server
with patch("hindsight_api.api.mcp.load_extension") as mock_load:
mock_load.return_value = None
create_mcp_server(mock_memory)
mock_load.assert_called_once_with("MCP", MCPExtension)
class TestMCPExtensionIntegration:
"""Integration tests verifying extension tools work end-to-end."""
@pytest.fixture
def mock_memory(self):
"""Create a mock MemoryEngine with required methods."""
memory = MagicMock()
memory.retain_batch_async = MagicMock()
memory.submit_async_retain = MagicMock(return_value={"operation_id": "test-op"})
memory.recall_async = MagicMock(return_value=MagicMock(results=[]))
memory.reflect_async = MagicMock(return_value=MagicMock(text="reflection"))
memory.list_banks = MagicMock(return_value=[])
memory.get_bank_profile = MagicMock(return_value={"id": "test"})
memory._tenant_extension = MagicMock()
return memory
def test_extension_tools_coexist_with_core_tools(self, mock_memory):
"""Extension tools are added alongside core tools, not replacing them."""
from hindsight_api.api.mcp import create_mcp_server
mock_ext = MockMCPExtension()
with patch("hindsight_api.api.mcp.load_extension", return_value=mock_ext):
mcp = create_mcp_server(mock_memory)
tools = mcp._tool_manager._tools
# All core tools present
assert "retain" in tools
assert "recall" in tools
assert "reflect" in tools
assert "list_banks" in tools
assert "create_bank" in tools
# Extension tool also present
assert "test_extension_tool" in tools
# At least 11 core + 1 extension = 12 tools (may grow as new tools are added)
assert len(tools) >= 12