Remove deno dependency in CI

.github/workflows/ci.yml

@@ -163,16 +163,10 @@ jobs:
           enable-cache: true
           cache-suffix: ${{ matrix.install.name }}
 
-      - uses: denoland/setup-deno@v2
-        with:
-          deno-version: v2.5.x
-
       - run: mkdir .coverage
 
       - run: uv sync --only-dev
 
-      - run: uv run mcp-run-python example --deps=numpy
-
       - name: cache HuggingFace models
         uses: actions/cache@v4
         with:
@@ -214,16 +208,10 @@ jobs:
           enable-cache: true
           cache-suffix: lowest-versions
 
-      - uses: denoland/setup-deno@v2
-        with:
-          deno-version: v2.5.x
-
       - run: mkdir .coverage
 
       - run: uv sync --group dev
 
-      - run: uv run mcp-run-python example --deps=numpy
-
       - name: cache HuggingFace models
         uses: actions/cache@v4
         with:

.github/workflows/claude.yml

@@ -39,10 +39,6 @@ jobs:
           enable-cache: true
           cache-suffix: claude-code
 
-      - uses: denoland/setup-deno@v2
-        with:
-          deno-version: v2.5.x
-
       - run: uv tool install pre-commit
 
       - run: make install

docs/contributing.md

@@ -9,26 +9,17 @@ git clone [email protected]:<your username>/pydantic-ai.git
 cd pydantic-ai
 ```
 
-Install `uv` (version 0.4.30 or later), `pre-commit` and `deno`:
+Install `uv` (version 0.4.30 or later) and `pre-commit`:
 
 - [`uv` install docs](https://docs.astral.sh/uv/getting-started/installation/)
 - [`pre-commit` install docs](https://pre-commit.com/#install)
-- [`deno` install docs](https://docs.deno.com/runtime/getting_started/installation/)
 
 To install `pre-commit` you can run the following command:
 
 ```bash
 uv tool install pre-commit
 ```
 
-For `deno`, you can run the following, or check
-[their documentation](https://docs.deno.com/runtime/getting_started/installation/) for alternative
-installation methods:
-
-```bash
-curl -fsSL https://deno.land/install.sh | sh
-```
-
 Install `pydantic-ai`, all dependencies and pre-commit hooks
 
 ```bash

docs/mcp/client.md

@@ -134,26 +134,22 @@ _(This example is complete, it can be run "as is" — you'll need to add `asynci
 
 MCP also offers [stdio transport](https://spec.modelcontextprotocol.io/specification/2024-11-05/basic/transports/#stdio) where the server is run as a subprocess and communicates with the client over `stdin` and `stdout`. In this case, you'd use the [`MCPServerStdio`][pydantic_ai.mcp.MCPServerStdio] class.
 
-In this example [mcp-run-python](https://github.com/pydantic/mcp-run-python) is used as the MCP server.
+In this example we use a simple MCP server that provides weather tools.
 
 ```python {title="mcp_stdio_client.py"}
 from pydantic_ai import Agent
 from pydantic_ai.mcp import MCPServerStdio
 
-server = MCPServerStdio(  # (1)!
-    'uv', args=['run', 'mcp-run-python', 'stdio'], timeout=10
-)
+server = MCPServerStdio('python', args=['mcp_server.py'], timeout=10)
 agent = Agent('openai:gpt-5', toolsets=[server])
 
 
 async def main():
-    result = await agent.run('How many days between 2000-01-01 and 2025-03-18?')
+    result = await agent.run('What is the weather in Paris?')
     print(result.output)
-    #> There are 9,208 days between January 1, 2000, and March 18, 2025.
+    #> The weather in Paris is sunny and 26 degrees Celsius.
 ```
 
-1. See [MCP Run Python](https://github.com/pydantic/mcp-run-python) for more information.
-
 ## Loading MCP Servers from Configuration
 
 Instead of creating MCP server instances individually in code, you can load multiple servers from a JSON configuration file using [`load_mcp_servers()`][pydantic_ai.mcp.load_mcp_servers].
@@ -167,9 +163,9 @@ The configuration file should be a JSON file with an `mcpServers` object contain
 ```json {title="mcp_config.json"}
 {
   "mcpServers": {
-    "python-runner": {
-      "command": "uv",
-      "args": ["run", "mcp-run-python", "stdio"]
+    "weather": {
+      "command": "python",
+      "args": ["mcp_server.py"]
     },
     "weather-api": {
       "url": "http://localhost:3001/sse"

docs/mcp/fastmcp-client.md

@@ -20,12 +20,12 @@ A `FastMCPToolset` can then be created from:
 
 - A FastMCP Server: `#!python FastMCPToolset(fastmcp.FastMCP('my_server'))`
 - A FastMCP Client: `#!python FastMCPToolset(fastmcp.Client(...))`
-- A FastMCP Transport: `#!python FastMCPToolset(fastmcp.StdioTransport(command='uvx', args=['mcp-run-python', 'stdio']))`
+- A FastMCP Transport: `#!python FastMCPToolset(fastmcp.StdioTransport(command='python', args=['mcp_server.py']))`
 - A Streamable HTTP URL: `#!python FastMCPToolset('http://localhost:8000/mcp')`
 - An HTTP SSE URL: `#!python FastMCPToolset('http://localhost:8000/sse')`
 - A Python Script: `#!python FastMCPToolset('my_server.py')`
 - A Node.js Script: `#!python FastMCPToolset('my_server.js')`
-- A JSON MCP Configuration: `#!python FastMCPToolset({'mcpServers': {'my_server': {'command': 'uvx', 'args': ['mcp-run-python', 'stdio']}}})`
+- A JSON MCP Configuration: `#!python FastMCPToolset({'mcpServers': {'my_server': {'command': 'python', 'args': ['mcp_server.py']}}})`
 
 If you already have a [FastMCP Server](https://gofastmcp.com/servers) in the same codebase as your Pydantic AI agent, you can create a `FastMCPToolset` directly from it and save agent a network round trip:
 
@@ -73,9 +73,9 @@ from pydantic_ai.toolsets.fastmcp import FastMCPToolset
 
 mcp_config = {
     'mcpServers': {
-        'time_mcp_server': {
-            'command': 'uvx',
-            'args': ['mcp-run-python', 'stdio']
+        'weather_server': {
+            'command': 'python',
+            'args': ['mcp_server.py']
         }
     }
 }

tests/example_modules/mcp_server.py

@@ -7,6 +7,12 @@
 mcp = FastMCP('Pydantic AI MCP Server')
 
 
+@mcp.tool()
+async def get_weather_forecast(location: str) -> str:
+    """Get the weather forecast for a location."""
+    return f'The weather in {location} is sunny and 26 degrees Celsius.'
+
+
 @mcp.tool()
 async def echo_deps(ctx: Context[ServerSessionT, LifespanContextT, RequestT]) -> dict[str, Any]:
     """Echo the run context.

tests/test_examples.py

@@ -301,6 +301,9 @@ def rich_prompt_ask(prompt: str, *_args: Any, **_kwargs: Any) -> str:
 
 
 class MockMCPServer(AbstractToolset[Any]):
+    def __init__(self, *args: Any, **kwargs: Any) -> None:
+        pass
+
     @property
     def id(self) -> str | None:
         return None  # pragma: no cover
@@ -336,6 +339,9 @@ async def call_tool(
     'What will the weather be like in Paris on Tuesday?': ToolCallPart(
         tool_name='weather_forecast', args={'location': 'Paris', 'forecast_date': '2030-01-01'}, tool_call_id='0001'
     ),
+    'What is the weather in Paris?': ToolCallPart(
+        tool_name='get_weather_forecast', args={'location': 'Paris'}, tool_call_id='0001'
+    ),
     'Tell me a joke.': 'Did you hear about the toothpaste scandal? They called it Colgate.',
     'Tell me a different joke.': 'No.',
     'Explain?': 'This is an excellent joke invented by Samuel Colvin, it needs no explanation.',
@@ -807,6 +813,8 @@ async def model_logic(  # noqa: C901
         return ModelResponse(parts=[TextPart('The current time is 10:45 PM on April 17, 2025.')])
     elif isinstance(m, ToolReturnPart) and m.tool_name == 'get_user':
         return ModelResponse(parts=[TextPart("The user's name is John.")])
+    elif isinstance(m, ToolReturnPart) and m.tool_name == 'get_weather_forecast':
+        return ModelResponse(parts=[TextPart(m.content)])
     elif isinstance(m, ToolReturnPart) and m.tool_name == 'get_company_logo':
         return ModelResponse(parts=[TextPart('The company name in the logo is "Pydantic."')])
     elif isinstance(m, ToolReturnPart) and m.tool_name == 'get_document':

tests/test_ui_web.py

@@ -19,6 +19,8 @@
     from pydantic_ai.builtin_tools import WebSearchTool
     from pydantic_ai.ui._web import create_web_app
 
+with try_import() as openai_import_successful:
+    import openai  # noqa: F401 # pyright: ignore[reportUnusedImport]
 
 pytestmark = [
     pytest.mark.skipif(not starlette_import_successful(), reason='starlette not installed'),
@@ -168,6 +170,7 @@ def test_chat_app_configure_endpoint_empty():
         )
 
 
+@pytest.mark.skipif(not openai_import_successful(), reason='openai not installed')
 def test_chat_app_configure_preserves_chat_vs_responses(monkeypatch: pytest.MonkeyPatch):
     """Test that openai-chat: and openai-responses: models are kept as separate entries."""
     monkeypatch.setenv('OPENAI_API_KEY', 'test-key')