github-repo-manager-mcp
This MCP server lets an AI client manage GitHub repositories through the GitHub REST and GraphQL APIs, enabling repository CRUD, topic/label management, and file operations.
README
GitHub Repo Manager MCP
This MCP server lets an AI client manage GitHub repositories through the GitHub REST and GraphQL APIs.
What This MCP Tool Can Do
- Create new repositories
- Update repo details, including description, homepage, topics/tags, and visibility
- Delete repositories
- Add topics/labels to repos for organization
- List all repos with filters
- Rename repositories
- Archive old repos
- Pin repositories via the GitHub GraphQL API
- Read file contents from repositories
- Create or update files in repositories, such as
README.md
Security Note
Never hard-code a GitHub token in source files. Set it as an environment variable named GITHUB_TOKEN.
If you pasted a real token into chat or committed it anywhere, revoke it in GitHub and create a new token.
Setup
npm install
Create a .env file in this project folder:
GITHUB_TOKEN=github_pat_your_token_here
The token needs permissions for the operations you want to use. For fine-grained personal access tokens, grant repository administration, contents, and metadata access for the target repositories.
Classic PAT notes:
- Deleting repositories requires the
delete_reposcope. - Updating repository files requires repo contents write access.
- Pinning repositories uses GitHub GraphQL
addPinnedItem; use a token that can access the target repository and update the authenticated user's pinned items.
Run
npm start
Install From npm
After this package is published, users can run it without cloning the repo:
npx github-repo-manager-mcp
Or install it globally:
npm install -g github-repo-manager-mcp
github-repo-manager-mcp
Set GITHUB_TOKEN in your MCP client environment. If you clone this repository locally, the server can also load GITHUB_TOKEN from a .env file in the project folder.
MCP Client Config
Example client config using npx:
{
"mcpServers": {
"github-repo-manager": {
"command": "npx",
"args": ["-y", "github-repo-manager-mcp"],
"env": {
"GITHUB_TOKEN": "github_pat_your_token_here"
}
}
}
}
Example local development config:
{
"mcpServers": {
"github-repo-manager": {
"command": "C:/PROGRA~1/nodejs/node.exe",
"args": ["C:/Users/msaad/OneDrive/Documents/github_organized/src/server.js"],
"env": {
"GITHUB_TOKEN": "github_pat_your_token_here"
}
}
}
}
The server also loads GITHUB_TOKEN from the local .env file automatically, so the env block is optional if .env is present.
Publish to npm
Before publishing, make sure you are logged in:
npm login
Then publish:
npm publish --access public
The package publishes only src, README.md, LICENSE, and run-mcp.cmd.
MCP Inspector on Windows
Use these settings in MCP Inspector:
Transport Type
STDIO
Command
C:\PROGRA~1\nodejs\node.exe
Arguments
C:/Users/msaad/OneDrive/Documents/github_organized/src/server.js
Use the short Node path to avoid the space in Program Files. Use forward slashes in the server path because MCP Inspector can strip backslashes from Windows paths in the Arguments field.
Available Tools
create_repository: Create a repository for the authenticated user or an organization. Supports description, homepage, privacy, auto-init,.gitignore, license, and topics.update_repository: Update description, homepage, visibility/privacy, feature flags, default branch, archived state, and topics.delete_repository: Delete a repository. Requiresconfirm=true.add_repository_topics: Add topics without removing existing topics.add_repository_labels: Create labels, or update existing labels if they already exist.list_repositories: List repositories visible to the authenticated user, an owner, or an organization with filters for visibility, archived state, topic, name, sort, and direction.rename_repository: Rename a repository.archive_repository: Archive a repository.pin_repository: Pin a repository to the authenticated user's GitHub profile via GraphQL.get_file_content: Read a file from a repository, defaulting toREADME.md.update_file_content: Create or update a file in a repository, defaulting toREADME.md; it auto-fetches the current SHA when updating an existing file.
Tool Inputs
Most tools accept an optional owner. If omitted, the server uses the authenticated GitHub user's login.
get_file_content inputs:
{
"owner": "optional-owner",
"repo": "repo-name",
"path": "README.md",
"branch": "optional-branch"
}
update_file_content inputs:
{
"owner": "optional-owner",
"repo": "repo-name",
"path": "README.md",
"content": "# New content",
"message": "Update README via MCP",
"branch": "optional-branch"
}
To update an existing file, sha is optional because the server attempts to fetch it automatically.
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.