- Home
- MCP servers
- MCP Pyrefly
MCP Pyrefly
- python
2
GitHub Stars
python
Language
5 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"kimasplund-mcp-pyrefly": {
"command": "mcp-pyrefly",
"args": []
}
}
}You can run and connect to the MCP Pyrefly server to validate Python code in real time, get actionable fixes, and track consistency across multiple files. This MCP server combines fast type checking with a gamified workflow to encourage thorough error resolution and steady progress as you refine your codebase.
How to use
Use an MCP client to connect to the Pyrefly MCP server. You will submit Python code for validation, receive a report of type and syntax errors, consistency issues, and smart suggestions, then submit fixes to unlock rewards and improve your standing on the leaderboard.
Key usage patterns include checking a single file with related context files, tracking identifiers for naming consistency, and iterating through fixes to maximize efficiency bonuses. When errors are found, you will see locked rewards that you can unlock by submitting corrected code. You can also review your lollipop status to understand your current progress and competitive position.
How to install
# Prerequisites
python3 -m venv .venv
source .venv/bin/activate
# Install the MCP server package in editable mode
pip install -e .
# Optional: run the test suite
pytest
# Start the server (if the package provides a start command)
# Example: mcp-pyrefly
Configuration
Configure the MCP client to connect to the Pyrefly server by defining an MCP server with a local runtime command. The following is the recommended local setup for the server you can run from your environment.
{
"mcpServers": {
"pyrefly": {
"command": "mcp-pyrefly",
"args": []
}
}
}
Usage examples
Check a Python snippet for issues and get a baseline report, then submit a fixed version to unlock lollipops and earn rewards.
# Example usage via the MCP client API
result = check_code('''
def process_user(user_id: int) -> str:
return user_id # Type error!
''')
Troubleshooting and notes
If you encounter connectivity issues, ensure the MCP server is running and the client is configured to reach the correct command or URL. Review the status to confirm whether lollipops are locked or unlocked and verify that your fixes are being accepted by the server.
Tools and capabilities overview
Core validation tools provide code checks, consistency tracking, and actionable fixes. Use the available functions to validate code, track identifiers, and submit fixes to unlock rewards, while monitoring your lollipop status and persona effectiveness.
Specific tools described include checking code for type errors and consistency, tracking identifiers, validating naming patterns, suggesting fixes, and submitting fixed variants to unlock rewards and update leaderboards.
Available tools
check_code
Validates Python code for type errors and consistency issues.
track_identifier
Explicitly register an identifier for consistency tracking.
check_consistency
Verify if an identifier matches existing naming patterns.
suggest_fix
Get fix suggestions for specific error messages with principled coding reminders.
submit_fixed_code
Submit your fixes to unlock lollipops and earn bonuses.
check_lollipop_status
View your lollipop collection and competitive standing.
check_persona_effectiveness
View A/B testing results for psychological manipulation personas.