MCPay Build
Enables no-code creation, preview, deployment, and monetization of MCP servers via x402 micropayments, including a virtual file system, sandbox preview, and tool monetization.
README
MCPay Build
<img src="./assets/mcpay-hero-painting.png" alt="MCPay Build Banner" width="600">
MCPay Build is an MCP server that acts as a no-code builder for MCP Servers with built-in monetization.
It lets developers design, preview, and deploy MCP servers to GitHub + Vercel β and instantly monetize tool usage with x402 micropayments via MCPay.
π Build β Preview β Deploy β Monetize
Our vision is an agent-to-agent economy powered by microtransactions, where AI agents can autonomously pay to use tools. MCPay Build provides the rails to make that future possible.
πΌοΈ Workflow
βββββββββββββ βββββββββββββ βββββββββββββ βββββββββββββββββ
β Build β ββββΆ β Preview β ββββΆ β Deploy β ββββΆ β Monetize β
β (AI/MCP) β β (sandbox) β β (Vercel) β β (MCPay/x402) β
βββββββββββββ βββββββββββββ βββββββββββββ βββββββββββββββββ
Big Picture
This repo implements an MCP server βfactoryβ:
- Lets an AI/chat (or you) spin up a brand-new MCP server in memory.
- Provides tools to edit files, manage deps, run type checks, format, version, and preview live.
- Deploys your server into a real Node.js environment using Vercel Sandbox.
- Monetizes expensive operations (
preview,run_command,stream_logs, etc.) using MCPay (per-call x402 micropayments).
Flow:
Create session β edit files β preview in sandbox β get live URL β iterate β export as ZIP β monetize tools.
β¨ Features
- MCPay integration β free + paid tools side-by-side, billed automatically
- Simple API β
server.tool(...)&server.paidTool(...)viamcpay/handler - Vercel-ready β
vercel.json+ function exports (GET/POST/OPTIONS) - Type-safe β modern TypeScript + Zod validation
- Dev UX β session/file/diff/format tools,
ts-check, logs, live preview
π Quickstart
1. Install dependencies
pnpm install
2. Configure environment
MCPAY_API_KEY=your_api_key # required for paid tools
MCPAY_API_URL=https://api.mcpay.tech # optional
3. Run locally
pnpm dlx vercel dev
Server runs at http://localhost:3000.
4. Connect from MCP client
Add to ~/.cursor/mcp.json:
{
"mcpServers": {
"mcpay-local": { "url": "http://localhost:3000" }
}
}
Under the Hood
Main runtime & routing
- Hono as the HTTP framework.
- Open CORS (all origins/headers/methods allowed).
GET /β serves a system prompt describing tools + philosophy.- All other routes β handled by
createPaidMcpHandler(...). - Function exports (
GET/POST/OPTIONS) β deploys cleanly to Vercel.
State & storage
- Sessions in Vercel KV (
session:{id}) with virtual file system + sandbox metadata. - Commits in KV (
commit:{session}:{commit}) for full snapshots. - Logs/ZIPs uploaded to Vercel Blob with public URLs.
Templates & defaults
When you run create_session, it seeds:
index.tsβ MCP server template (withhello+get-timetools).package.json,tsconfig.json,.gitignore,.env.example,README.md.
Preview = live sandbox
preview tool does:
-
Loads session + hashes files.
-
Reuses existing sandbox if files unchanged (<30 min).
-
Else:
- Creates fresh Vercel Sandbox (Node 22, 2 vCPUs).
- Uploads files.
- Detects Node/npm versions.
- Installs deps (npm/yarn/pnpm).
- Runs build if available.
- Starts server (
npm startor fallback). - Health-checks endpoints until ready.
- Aggregates logs (uploads long logs to Blob).
Returns: sandbox ID, preview URL, logs, node/npm versions, status.
Built-in Tools
Paid (metered via MCPay)
create_sessionβ bootstrap projectpreviewβ deploy to sandbox, return live URLdownload_sessionβ zip + upload projectrun_commandβ exec arbitrary shell in sandboxstream_logsβ live-tail logs
Free
- File ops:
add_or_update_file,delete_file,get_file,list_files,get_all_codebase - Env ops:
add_env_key,list_env_keys,delete_env_key - Package ops:
add_dependency,debug_package_json - Quality:
ts_check,npm_audit,format_file - Versioning:
commit,diff,revert,get_logs - Info:
get_environment_info,get_session_stats,stop_preview,force_refresh_sandbox
Monetization
- MCP handler is created with
createPaidMcpHandler. - Paid tools have explicit
{ price, currency }. - Requires
MCPAY_API_KEY(and optionalMCPAY_API_URL).
Example:
server.paidTool(
"analyze-data",
"Analyze data (paid)",
{ price: 0.1, currency: "USD" },
{ data: z.string(), format: z.enum(["json","csv"]) },
async ({ data, format }) => ({
content: [{ type: "text", text: `Analyzed ${format} data` }]
})
);
π Typical Flow
create_sessionβ new IDadd_or_update_fileβ editindex.tsts_check+format_filecommitβ snapshotpreviewβ live sandbox URLstream_logsβ monitor in real timedownload_sessionβ zip/export
π¦ Deployment
Deploy to Vercel:
pnpm dlx vercel
Set env vars in project settings:
MCPAY_API_KEYMCPAY_API_URL(optional)
β Notes
- Blob uploads are public β donβt commit secrets.
- Sandbox per tool (for some ops) β isolation at cost of speed/$$.
- .env handling β only keys are tracked; actual secrets must be set in runtime env.
π License
MIT
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.