Yellhorn MCP

A Model Context Protocol (MCP) server that exposes Gemini 2.5 Pro capabilities to Claude Code for software development tasks using your entire codebase in the prompt. This pattern is highly useful for defining work to be done by code assistants like Claude Code or other MCP compatible coding agents, and reviewing the results ensuring they meet the exactly specified original requirements.
Features
- Create Workplans: Creates detailed implementation plans based on a prompt and taking into consideration your entire codebase, posting them as GitHub issues and exposing them as MCP resources for your coding agent
- Review Code Diffs: Provides a tool to evaluate git diffs against the original workplan with full codebase context and provides detailed feedback, ensuring the implementation does not deviate from the original requirements and providing guidance on what to change to do so
- Isolated Development Environments: Creates Git worktrees and linked branches for streamlined, isolated development workflow (can be done separately from workplan generation), allowing parallel development with multiple agents
- Seamless GitHub Integration: Automatically creates labeled issues with proper branch linking in the GitHub UI, posts reviews subissues with references to original workplan issues.
- Context Control: Use
.yellhornignorefiles to exclude specific files and directories from the AI context, similar to.gitignore - MCP Resources: Exposes workplans as standard MCP resources for easy listing and retrieval
Installation
# Install from PyPI
pip install yellhorn-mcp
# Install from source
git clone https://github.com/msnidal/yellhorn-mcp.git
cd yellhorn-mcp
pip install -e .
Configuration
The server requires the following environment variables:
GEMINI_API_KEY: Your Gemini API key (required)REPO_PATH: Path to your repository (defaults to current directory)YELLHORN_MCP_MODEL: Gemini model to use (defaults to “gemini-2.5-pro-exp-03-25”)
The server also requires the GitHub CLI (gh) to be installed and authenticated.
Usage
Running the server
# As a standalone server
yellhorn-mcp --repo-path /path/to/repo --host 127.0.0.1 --port 8000
# Using the MCP CLI
mcp dev yellhorn_mcp.server
# Install as a permanent MCP server for Claude Desktop
mcp install yellhorn_mcp.server
# Set environment variables during installation
mcp install yellhorn_mcp.server -v GEMINI_API_KEY=your_key_here -v REPO_PATH=/path/to/repo
Integration with Claude Code
When working with Claude Code, you can use the Yellhorn MCP tools by:
Starting a project task:
Please generate a workplan with title "[Your Title]" and detailed description "[Your detailed requirements]"Create a worktree for the workplan (optional):
Please create a worktree for issue #123Navigate to the created worktree directory:
cd [worktree_path] # The path is returned in the responseView the workplan if needed:
# You can run this from anywhere Please get the workplan for issue #123Make your changes, create a PR, and request a review:
# First create a PR using your preferred method (Git CLI, GitHub CLI, or web UI) git add . git commit -m "Implement feature" git push origin HEAD gh pr create --title "[PR Title]" --body "[PR Description]" # You can run this from anywhere Please review the PR comparing "main" and "feature-branch" against the workplan in issue #123
Tools
create_workplan
Creates a GitHub issue with a detailed workplan based on the title and detailed description.
Input:
title: Title for the GitHub issue (will be used as issue title and header)detailed_description: Detailed description for the workplan
Output:
- JSON string containing:
issue_url: URL to the created GitHub issueissue_number: The GitHub issue number
create_worktree
Creates a git worktree with a linked branch for isolated development from an existing workplan issue.
Input:
issue_number: The GitHub issue number for the workplan
Output:
- JSON string containing:
worktree_path: Path to the created Git worktree directorybranch_name: Name of the branch created for the worktreeissue_url: URL to the associated GitHub issue
get_workplan
Retrieves the workplan content (GitHub issue body) associated with a workplan.
Input:
issue_number: The GitHub issue number for the workplan.
Output:
- The content of the workplan issue as a string
review_workplan
Triggers an asynchronous code review comparing two git refs (branches or commits) against a workplan described in a GitHub issue. Creates a GitHub sub-issue with the review asynchronously after running (in the background).
Input:
issue_number: The GitHub issue number for the workplan.base_ref: Base Git ref (commit SHA, branch name, tag) for comparison. Defaults to ‘main’.head_ref: Head Git ref (commit SHA, branch name, tag) for comparison. Defaults to ‘HEAD’.
Output:
- A confirmation message that the review task has been initiated
Resource Access
Yellhorn MCP also implements the standard MCP resource API to provide access to workplans:
list-resources: Lists all workplans (GitHub issues with the yellhorn-mcp label)get-resource: Retrieves the content of a specific workplan by issue number
These can be accessed via the standard MCP CLI commands:
# List all workplans
mcp list-resources yellhorn-mcp
# Get a specific workplan by issue number
mcp get-resource yellhorn-mcp 123
Development
# Install development dependencies
pip install -e ".[dev]"
# Run tests
pytest
CI/CD
The project uses GitHub Actions for continuous integration and deployment:
Testing: Runs automatically on pull requests and pushes to the main branch
- Linting with flake8
- Format checking with black
- Testing with pytest
Publishing: Automatically publishes to PyPI when a version tag is pushed
- Tag must match the version in pyproject.toml (e.g., v0.2.2)
- Requires a PyPI API token stored as a GitHub repository secret (PYPI_API_TOKEN)
To release a new version:
- Update version in pyproject.toml
- Commit changes:
git commit -am "Bump version to X.Y.Z" - Tag the commit:
git tag vX.Y.Z - Push changes and tag:
git push && git push --tags
For more detailed instructions, see the Usage Guide.
License
MIT
Yellhorn
Project Details
- msnidal/yellhorn-mcp
- Last Updated: 4/12/2025
Recomended MCP Servers
A TypeScript-based MCP server for Jira integration with Cursor
A declarative, composable model context protocol server that uses a remote config
一个基于MCP协议的搜索服务实现,提供网络搜索和本地搜索功能,Cursor和Claude Desktop能与之无缝集成。
Pre-trained Deep Learning models and samples (high quality and extremely fast)
Stock market data provider for Claude Desktop using MCP
MCP server for shadcn/ui component references
Netflix-level subtitle cutting, translation, alignment, and even dubbing - one-click fully automated AI video subtitle team | Netflix级字幕切割、翻译、对齐、甚至加上配音,一键全自动视频搬运AI字幕组
Deep Research MCP is an intelligent research assistant built on the Model Context Protocol (MCP) that performs comprehensive,...





