
MCP: Local PR Description Generator
A lightweight tool that uses Mistral AI running locally to generate structured GitHub pull request descriptions from git diffs.
README
🧠 MCP:Local PR Description Generator using Mistral
This project is a lightweight tool that uses Mistral running locally via Ollama to generate clean, structured GitHub pull request (PR) descriptions from staged git diffs.
It wraps the model inside a FastAPI server and provides a Gradio web UI for quick generation.
📌 Features
- 💻 Local-first: All processing is done on your machine using
ollama
(no cloud calls) - ⚙️ FastAPI backend: Exposes a simple HTTP endpoint to query Mistral
- 🌐 Gradio frontend: Paste a git diff and get a structured PR description instantly
- 🧩 Git-aware: Designed around generating summaries from actual diffs
- 🪶 Lightweight and privacy-friendly
🚀 Getting Started
1. Install Dependencies
pip install -r requirements.txt
2. Start Ollama with Mistral
ollama run mistral
3. Run the FastAPI server
uvicorn app.server:app --reload
4. Start the Gradio UI
python -m frontend.gradio
✨ Example
Git Diff
diff --git a/src/utils.py b/src/utils.py
index 1a2b3c4..5d6ef7f 100644
--- a/src/utils.py
+++ b/src/utils.py
@@ def calculate_average(numbers):
- return sum(numbers) / len(numbers)
+ if not numbers:
+ return 0
+ return sum(numbers) / len(numbers)
@@ def greet_user(name):
- print("Hello", name)
+ print(f"Hello, {name} 👋")
Output PR Description
**Title:** [Utils] Handle empty list in average calculation and improved greeting
**Description:**
- Added edge case handling for `calculate_average()` to return 0 when given an empty list.
- Enhanced user experience by adding an emoji in the greeting function `greet_user()`.
These changes improve error handling and make output more user-friendly.
🧠 How it Works
- The Gradio UI sends the pasted diff to
http://localhost:8000/mistral
- FastAPI receives it and forms a prompt for the local Mistral model
- Mistral responds with a human-readable PR title + description
- The Gradio UI displays the result
🔒 Privacy Note
This tool is completely local. Nothing is sent to the cloud — your code and git diffs stay on your machine.
🙌 Credits
📬 License
MIT License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.