x402 Stellar MCP Server
Enables AI agents to pay for protected HTTP resources using Stellar USDC via the x402 protocol, facilitating automated payments and access to paid APIs.
README
x402 Stellar MCP Server
Local MCP server (stdio) that can call x402-protected HTTP resources and automatically pay in Stellar USDC.
Features
- Payment flow (
stellar:testnetorstellar:pubnet) - Default testnet configuration
- Mainnet-ready via env switch
- Optional OpenZeppelin facilitator auth via env
- No resource whitelist: URL is provided at tool call time
Tools Exposed
x402_wallet_info: shows active wallet address/network/configx402_facilitator_supported: checks configured facilitator/supportedendpointfetch_paid_resource: generic paid fetch (url,method, optional body/headers)
Prerequisites
- Node.js 20+
- A Stellar wallet funded with testnet XLM + testnet USDC
- x402 resource server running (your
app/index.jsdemo) - x402 facilitator running on
http://localhost:4022
Setup
npm install
cp .env.example .env
Testnet
For testnet you'll need to run a facilitator locally on port 4022. I'd recommend using the one included in the coinbase x402 repo: https://github.com/coinbase/x402/tree/main/examples/typescript/facilitator/advanced
Update .env with your wallet key:
STELLAR_SECRET_KEY=S...
STELLAR_NETWORK=stellar:testnet
Pubnet
For mainnet you'll just need an OpenZeppelin API Key for the relayer. You can get one here (Thank you for making this easy): https://channels.openzeppelin.com/gen
STELLAR_NETWORK=stellar:pubnet
STELLAR_RPC_URL=https://rpc.lightsail.network/
X402_FACILITATOR_URL=https://channels.openzeppelin.com/x402
X402_FACILITATOR_API_KEY=<your-openzeppelin-api-key>
X402_FACILITATOR_API_KEY is optional in general, but required for OpenZeppelin channels.
When set, fetch_paid_resource auto-adds Authorization: Bearer <key> for requests under X402_FACILITATOR_URL (unless you provide Authorization explicitly).
Run
npm run dev
The MCP server runs over stdio for Claude/Codex integrations.
TTS Proxy
The paid /api/tts route can use either elevenlabs or smallest.
Set a default provider in .env, and optionally override it per request by sending "provider": "elevenlabs" or "provider": "smallest" in the JSON body.
You can also send "provider": "auto" with a routing "priority" of "cheap", "fast", or "quality" to let the server pick a provider dynamically.
To run the real x402-protected TTS proxy, configure:
STELLAR_NETWORK=stellar:testnet
X402_FACILITATOR_URL=http://localhost:4022
TTS_PROVIDER=elevenlabs
ELEVENLABS_API_KEY=<your-elevenlabs-api-key>
ELEVENLABS_MODEL_ID=eleven_flash_v2_5
ELEVENLABS_VOICE_ID=21m00Tcm4TlvDq8ikWAM
ELEVENLABS_ESTIMATED_PRICE_USD=0.06
ELEVENLABS_ESTIMATED_LATENCY_MS=900
ELEVENLABS_QUALITY_SCORE=9.5
SMALLEST_API_KEY=<your-smallest-api-key>
SMALLEST_MODEL=lightning
SMALLEST_VOICE_ID=emily
SMALLEST_LANGUAGE=en
SMALLEST_ESTIMATED_PRICE_USD=0.03
SMALLEST_ESTIMATED_LATENCY_MS=500
SMALLEST_QUALITY_SCORE=8.3
ALLOW_SYSTEM_PROXY=false
PAY_TO_ADDRESS=<your-stellar-receiving-address>
For pubnet, also set:
STELLAR_NETWORK=stellar:pubnet
STELLAR_RPC_URL=https://rpc.lightsail.network/
X402_FACILITATOR_URL=https://channels.openzeppelin.com/x402
X402_FACILITATOR_API_KEY=<your-openzeppelin-api-key>
Then start the proxy with:
npm run start:elevenlabs
The protected route is:
POST http://localhost:3000/api/tts
Example request for automatic routing:
{
"text": "Hello from the voice marketplace",
"provider": "auto",
"priority": "cheap",
"language": "en"
}
The response includes marketplace metadata headers like:
X-TTS-Provider
X-TTS-Priority
X-TTS-Reason
X-TTS-Estimated-Price-Usd
X-TTS-Estimated-Latency-Ms
X-TTS-Quality-Score
This server now requires a real facilitator and will not fall back to a mock implementation.
By default it also ignores HTTP_PROXY / HTTPS_PROXY / ALL_PROXY so upstream TTS requests go direct during local testing. Set ALLOW_SYSTEM_PROXY=true if you intentionally want to use a proxy.
Add To Claude / Codex
Use an MCP entry like (don't forget to change the path/):
Codex
codex mcp add x402-stellar -- npm --silent --prefix /absolute/path/to/x402-mcp-stellar run dev
Claude Code
claude mcp add x402-stellar -- npm --silent --prefix /absolute/path/to/x402-mcp-stellar run dev
Claude Desktop (untested)
{
"mcpServers": {
"x402-stellar": {
"command": "npm",
"args": ["--silent", "--prefix", "/absolute/path/to/x402-mcp-stellar", "run", "dev"]
}
}
}
This server loads .env from the project directory (/absolute/path/to/x402-mcp-stellar/.env).
Claude Usage
After loading the MCP server, you can ask:
Can you fetch the resource at http://localhost:3000/my-service, use the x402-stellar MCP server to pay for it, and print the response?
The tool call will pass that full URL at runtime; no URL allowlist or hardcoded endpoint is required in this MCP server.
Notes
@x402/stellaris not currently published on npm, so this repo vendors the Stellar mechanism undersrc/stellar.- Default testnet Soroban RPC is used automatically.
- Mainnet requires
STELLAR_RPC_URL.
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.