feat: Allow any value type in get_prompt arguments

The type hints for the `arguments` parameter in `ClientSession.get_prompt` and
the corresponding `GetPromptRequestParams` were previously restricted to `dict[str, str]`.

This was overly restrictive and prevented passing arguments with non-string values,
such as numbers or booleans. This commit changes the type to `dict[str, Any]`, allowing
for more flexible prompt argument structures.

A corresponding test case has been added to verify that calling `get_prompt` with an integer
argument now works as expected.

Fixes #749

Author: arjunaskykok

src/mcp/client/session.py

@@ -344,7 +344,7 @@ async def list_prompts(self, cursor: str | None = None) -> types.ListPromptsResu
             types.ListPromptsResult,
         )
 
-    async def get_prompt(self, name: str, arguments: dict[str, str] | None = None) -> types.GetPromptResult:
+    async def get_prompt(self, name: str, arguments: dict[str, Any] | None = None) -> types.GetPromptResult:
         """Send a prompts/get request."""
         return await self.send_request(
             types.ClientRequest(

src/mcp/types.py

@@ -639,7 +639,7 @@ class GetPromptRequestParams(RequestParams):
 
     name: str
     """The name of the prompt or prompt template."""
-    arguments: dict[str, str] | None = None
+    arguments: dict[str, Any] | None = None
     """Arguments to use for templating the prompt."""
     model_config = ConfigDict(extra="allow")
 

tests/client/test_session.py

@@ -495,3 +495,72 @@ async def mock_server():
     assert received_capabilities.roots is not None  # Custom list_roots callback provided
     assert isinstance(received_capabilities.roots, types.RootsCapability)
     assert received_capabilities.roots.listChanged is True  # Should be True for custom callback
+
+
+@pytest.mark.anyio
+async def test_get_prompt_with_non_string_argument():
+    client_to_server_send, client_to_server_receive = anyio.create_memory_object_stream[SessionMessage](1)
+    server_to_client_send, server_to_client_receive = anyio.create_memory_object_stream[SessionMessage](1)
+
+    async def mock_server():
+        await client_to_server_receive.receive()
+
+        await server_to_client_send.send(
+            SessionMessage(
+                JSONRPCMessage(
+                    JSONRPCResponse(
+                        jsonrpc="2.0",
+                        id=0,
+                        result=InitializeResult(
+                            protocolVersion=LATEST_PROTOCOL_VERSION,
+                            capabilities=ServerCapabilities(),
+                            serverInfo=Implementation(name="mock-server", version="0.1.0"),
+                        ).model_dump(by_alias=True, mode="json", exclude_none=True),
+                    )
+                )
+            )
+        )
+
+        await client_to_server_receive.receive()
+
+        # Receive get_prompt request
+        session_message = await client_to_server_receive.receive()
+        jsonrpc_request = session_message.message
+        assert isinstance(jsonrpc_request.root, JSONRPCRequest)
+        request = ClientRequest.model_validate(
+            jsonrpc_request.model_dump(by_alias=True, mode="json", exclude_none=True)
+        )
+        assert isinstance(request.root, types.GetPromptRequest)
+        assert request.root.params.arguments == {"employee_id": 77}
+
+        # Send get_prompt result
+        await server_to_client_send.send(
+            SessionMessage(
+                JSONRPCMessage(
+                    JSONRPCResponse(
+                        jsonrpc="2.0",
+                        id=jsonrpc_request.root.id,
+                        result=types.GetPromptResult(
+                            messages=[
+                                types.PromptMessage(role="user", content=types.TextContent(type="text", text="..."))
+                            ]
+                        ).model_dump(by_alias=True, mode="json", exclude_none=True),
+                    )
+                )
+            )
+        )
+
+    async with (
+        ClientSession(
+            server_to_client_receive,
+            client_to_server_send,
+        ) as session,
+        anyio.create_task_group() as tg,
+        client_to_server_send,
+        client_to_server_receive,
+        server_to_client_send,
+        server_to_client_receive,
+    ):
+        tg.start_soon(mock_server)
+        await session.initialize()
+        await session.get_prompt("get_employee_profile", {"employee_id": 77})