Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@python-runner run Python code to calculate the first 10 prime numbers"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP (Model Context Protocol) server that exposes a run_script tool for executing Python code via subprocess.
uv syncRelated MCP server: Python Executor MCP Server
uv run python server.pyclaude mcp add --transport stdio --scope user python-runner -- uvx --from git+https://github.com/Drahoxx/python-mcp python-mcpOr for a local installation:
claude mcp add --transport stdio --scope user python-runner -- uv run --directory /path/to/python-mcp python-mcpAdd to ~/.config/opencode/opencode.json:
{
"mcpServers": {
"python-runner": {
"command": "uvx",
"args": ["--from", "git+https://github.com/Drahoxx/python-mcp", "python-mcp"]
}
}
}Or for a local installation:
{
"mcpServers": {
"python-runner": {
"command": "uv",
"args": ["run", "--directory", "/path/to/python-mcp", "python-mcp"]
}
}
}Add to your MCP client configuration:
{
"mcpServers": {
"python-runner": {
"command": "uv",
"args": ["run", "--directory", "/path/to/python-mcp", "python", "server.py"]
}
}
}run_scriptExecute Python code and return the result.
Name | Type | Default | Description |
|
| required | Python code to execute |
|
|
| Execution timeout in seconds |
Returns a JSON string with:
Field | Type | Description |
|
| Whether execution succeeded (exit code 0) |
|
| Captured standard output |
|
| Captured standard error |
|
| Error message/traceback if failed |
|
| Process exit code (-1 for timeout) |
Successful execution:
run_script('print("hello")')
# {"success": true, "stdout": "hello\n", "stderr": "", "error": null, "return_code": 0}Syntax error:
run_script('print(')
# {"success": false, "stdout": "", "stderr": "...", "error": "SyntaxError...", "return_code": 1}Runtime error:
run_script('1/0')
# {"success": false, "stdout": "", "stderr": "...", "error": "ZeroDivisionError...", "return_code": 1}Timeout:
run_script('import time; time.sleep(60)', timeout=2)
# {"success": false, "stdout": "", "stderr": "", "error": "Execution timed out after 2 seconds", "return_code": -1}uv run pytestuv run pytest --cov=serverExecute Python code and return the result.
Args: code: Python code to execute timeout: Execution timeout in seconds (default: 30)
Returns: JSON string with success, stdout, stderr, error, and return_code
Inputs: code (required), timeout. Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 1 tool
Only one tool exists, so there is no possibility of confusion between tools. The tool's purpose is clearly defined and unambiguous.
The single tool uses a consistent verb_noun naming convention (run_script), which is clear and predictable.
The server has exactly one tool, which is slightly below the typical 3-15 range, but it directly matches the server's stated purpose of running Python code. It could potentially benefit from additional tools like environment management, but the core functionality is well-covered.
The tool covers the full lifecycle of executing a Python script, including timeout and capturing stdout/stderr/errors. For the narrow domain of running Python code, it is complete.
Run Python code from natural language prompts, with optional PyPI package install.
Run sandboxed Python for calculations and PDFs; no other languages, installs or network.
Hosted one-shot code execution: send a snippet plus optional files, get stdout, stderr, exit...
Run Python code in a secure sandbox without local setup. Declare inline dependencies and execute s…
