mcp-app
Fetches public GitHub repositories and uses an LLM to generate a json-render UI specification.
README
GitHub MCP App Demo (json-render + LLM)
An MCP App that fetches public GitHub repositories and uses an LLM to generate a json-render UI spec.
Project layout
mcp-app/
├── server/ # MCP server (tools, GitHub API, LLM spec generation)
│ ├── main.ts # entry — HTTP / stdio transport
│ ├── server.ts # tool + resource registration
│ ├── github.ts # GitHub REST API
│ ├── llm-spec.ts # DeepSeek / OpenAI → json-render spec
│ ├── catalog.ts # json-render component catalog
│ └── types.ts
├── app/ # MCP App UI (iframe)
│ ├── index.html
│ ├── main.tsx
│ └── registry.tsx
├── dist/ # build output (index.html bundle)
└── .env
Flow
GitHub API → repos data → LLM + catalog.prompt() → json-render Spec → Renderer
Setup
npm install
cp .env.example .env
# Set DEEPSEEK_API_KEY and DEEPSEEK_MODEL in .env
npm run build
npm start
Server listens at http://localhost:3001/mcp.
Test with basic-host
cd /Users/CYKUO/Codes/ext-apps/examples/basic-host && npm start
Open http://localhost:8080 → http://localhost:3001/mcp → pick a tool → Call Tool.
Tools
All tools require DEEPSEEK_API_KEY and use the LLM to generate the UI spec. The layout parameter (table or cards) is passed as a hint to the LLM.
list-repos
| Parameter | Description |
|---|---|
username |
GitHub user (default: modelcontextprotocol) |
perPage |
Max repos to fetch (default: 10) |
layout |
table or cards (default: table) |
prompt |
Natural language UI instructions for the LLM |
get-user-profile
| Parameter | Description |
|---|---|
username |
GitHub user (default: modelcontextprotocol) |
layout |
table or cards (default: table) |
prompt |
Natural language UI instructions for the LLM |
Returns profile fields such as bio, followers, following, public repo count, location, and company.
list-repo-issues
| Parameter | Description |
|---|---|
owner |
Repo owner (default: modelcontextprotocol) |
repo |
Repo name (default: ext-apps) |
state |
open, closed, or all (default: open) |
perPage |
Max issues to fetch (default: 10) |
layout |
table or cards (default: table) |
prompt |
Natural language UI instructions for the LLM |
Pull requests are filtered out; only issues are returned.
Environment Variables
| Variable | Required | Description |
|---|---|---|
DEEPSEEK_API_KEY |
Yes | DeepSeek API key |
DEEPSEEK_MODEL |
No | e.g. deepseek-v4-flash |
DEEPSEEK_BASE_URL |
No | Default https://api.deepseek.com/v1 |
GITHUB_TOKEN |
No | Higher GitHub rate limits |
PORT |
No | Default 3001 |
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
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
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.