About GitHub Project Manager MCP
GitHub Project Manager MCP is an MCP server for managing GitHub resources through the GitHub CLI (`gh`). It focuses on issues and pull requests, with GitHub Projects integration listed as future work.
Key features
- 01Manage GitHub issues
- 02Manage pull requests
- 03Use GitHub CLI (`gh`)
- 04Run as a Docker container
- 05Integration tests against the GitHub API
Use cases
- 01Manage issues in a GitHub repository
- 02Work with pull requests through MCP tools
- 03Run the server with a GitHub token in Docker
- 04Test GitHub operations against a test repository
MCP support
Exposes GitHub management tools to MCP clients, using the GitHub CLI (`gh`) for operations. The README describes configuring it in Cursor.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
docker run -i --rm -e GH_TOKEN ghcr.io/marioalvial/gh-project-manager-mcpGH_TOKENSecretLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
ghcr.io/marioalvial/gh-project-manager-mcpstdioTools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEGitHub Project Manager MCP
Um servidor MCP (Model Context Protocol) para gerenciar projetos no GitHub.
Descrição
Este projeto implementa um servidor MCP especializado em operações de gerenciamento de projetos GitHub, oferecendo integração com o GitHub CLI (gh) para tarefas comuns como:
- Gerenciamento de issues
- Gerenciamento de pull requests
- Futuramente: Integração com GitHub Projects
Uso com Docker
A imagem Docker é publicada automaticamente no GitHub Container Registry a cada release:
docker run -i --rm -e GH_TOKEN=seu_token ghcr.io/marioalvial/gh-project-manager-mcp
Configuração do MCP em Cursor
Para usar este MCP com o Cursor, adicione a seguinte configuração ao seu .cursor/mcp.json:
"gh-project-manager": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"GH_TOKEN",
"ghcr.io/marioalvial/gh-project-manager-mcp"
],
"env": {
"GH_TOKEN": "seu_token_github"
}
}
Versionamento
Este projeto usa Semantic Versioning e libera automaticamente novas versões baseadas nos padrões de commit.
Integration Tests
This project includes a comprehensive suite of integration tests that verify the functionality of our GitHub MCP tools with the actual GitHub API. These tests create and manage real GitHub resources during test execution and clean up afterward.
Running Integration Tests
To run the integration tests, you'll need a GitHub token with appropriate permissions:
- Create a GitHub personal access token with at least these scopes:
repo,project - Create a
.envfile in the project root with the following content:GH_INTEGRATION_TEST_TOKEN=your_github_token_here GH_INTEGRATION_TEST_OWNER=marioalvial # or your own username/organization GH_INTEGRATION_TEST_REPO=gh-project-manager-mcp # or your own repository GH_INTEGRATION_TEST_PROJECT_ID=your_project_id # required for project tests - Run the integration tests:
make function-integration-test
Test Approach
The integration tests use direct function calls to the implementation functions, bypassing the MCP protocol layer to focus on the core GitHub CLI interactions. This approach ensures that:
- Our tool implementations correctly interact with the GitHub CLI commands
- Resource creation and cleanup work as expected
- Error handling is robust across various scenarios, including token permission limitations
The tests are designed to handle various GitHub API limitations and token permission scenarios, automatically skipping or adapting tests that require permissions not available to the provided token.
Note: The tests create and manage real GitHub resources (issues, PRs, branches, etc.), so make sure to use a test repository to avoid clutter in production repositories.
Test line for pull request testing - update for PR testing.
Demo PR
This change demonstrates creating a PR using the MCP tools.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.