GitHub Repo Assistant MCP Server
Enables creating GitHub repositories under a fixed owner and pushing multiple files in a single commit via the Git Data API, designed for deployment as a Vercel serverless function.
README
GitHub Repo Assistant (MCP server)
An MCP server with two tools:
- create_repo — creates a GitHub repo under a fixed owner if it doesn't already exist
- push_files — creates/updates one or more files in a single commit (creates the branch if needed)
Designed to run as a Vercel serverless function so Claude can call it as a remote MCP connector.
1. Push this code to GitHub
cd github-repo-assistant-mcp
git init
git add .
git commit -m "Initial commit: github repo assistant MCP server"
git branch -M main
git remote add origin https://github.com/<your-username>/github-repo-assistant-mcp.git
git push -u origin main
2. Deploy to Vercel
-
Go to https://vercel.com/new and import the GitHub repo you just pushed.
-
Leave build settings as default (no build step needed — it's a single API route).
-
Before deploying (or right after, then redeploy), add these Environment Variables in the Vercel project settings:
Name Value GITHUB_TOKENA GitHub Personal Access Token with reposcope (fine-grained: Contents + Administration read/write)GITHUB_OWNERYour GitHub username or org name — the account repos will be created under -
Deploy. Your MCP endpoint will be:
https://<your-project-name>.vercel.app/api/mcp
3. Connect it to Claude
Add it as a custom connector using that URL. Once connected, you can ask Claude things like:
"Create a repo called
weekend-projectand push a basic Express server to it."
Claude will call create_repo then push_files against your GitHub account.
Creating a GitHub token
- GitHub → Settings → Developer settings → Personal access tokens → Fine-grained tokens
- Repository access: "All repositories" (or select ones, but then new repos can't be created into arbitrary names — "All repositories" is simpler for this use case)
- Permissions needed: Contents: Read and write, Administration: Read and write (Administration is required to create new repos)
Notes
- This server is stateless — each tool call is an independent request to the GitHub REST/Git Data API, which fits Vercel's serverless model well.
push_filesuses the Git Data API (trees/commits/refs) so multiple files land in a single commit rather than one API call per file.- All repos are created under the single
GITHUB_OWNERconfigured via environment variable. If you want Claude to create repos under different owners/orgs per request, add an optionalownerparameter to the tools and pass it through instead of reading fromprocess.env.
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.