peage-mcp
MCP server that provides AI agents a prepaid fiat wallet, enabling them to pay metered APIs per call on the peage rail with adjustable spending caps.
README
peage-mcp — give your agent a wallet
An MCP server that puts a prepaid fiat wallet in any MCP-capable agent (Claude Code, Claude Desktop, anything speaking MCP stdio). The agent can then pay APIs per call on the peage rail — cents at a time, no crypto, no subscriptions, with spending caps you control.
One ~120 KB static binary, pure machin (MFL). No Node, no Python.
Install
Grab the binary (or ./build.sh with machin on PATH), then register it:
Claude Code (.mcp.json in your project, or ~/.claude.json for user scope):
{"mcpServers": {"peage": {"command": "/path/to/peage-mcp", "args": ["serve"]}}}
Claude Desktop (claude_desktop_config.json): same snippet under mcpServers.
What the agent gets (7 tools)
| tool | what it does |
|---|---|
peage_setup |
mint the wallet once; token stored at ~/.peage-mcp.json |
peage_status |
balance, escrow, caps, recent charges |
peage_topup |
Stripe Checkout URL → the agent hands it to you, you pay once |
peage_paid_request |
call any peage-metered API with the wallet attached |
peage_set_limits |
per-call + per-merchant daily spending caps |
peage_verify_receipt |
publicly verify any signed charge receipt |
peage_solvency |
audit the rail's live ledger invariant |
The human appears exactly once in the loop: paying the top-up link. Everything else is the agent's business.
The 60-second demo
Ask your agent: "set up a peage wallet, then fetch https://peage.intrane.fr/demo/fortune
as a paid request" — it mints a wallet that comes with free starter credit, so it pays
the 1-cent toll and gets its fortune on the first try, no human needed. When the free
credit runs out, peage_topup hands you a Stripe link to fund more. That's the whole
agent-payment loop — and the agent completes it before you touch your card.
Security model
- The wallet token lives in
~/.peage-mcp.json(orPEAGE_WALLET_TOKEN). Sharing it with a merchant API is how they charge you — the caps bound the exposure: default 1€ per call / 10€ per merchant per day, adjustable viapeage_set_limits. - The rail stores only a hash of the token; every charge returns an HMAC-signed receipt
anyone can verify; the ledger invariant is public (
peage_solvency). PEAGE_URLpoints the server at a different rail instance (self-hosted, staging).
Sibling repos: peage (the rail, docs, skills) · x402: peage is a live x402 facilitator. An intrane.fr product. Contact: javi@intrane.fr
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.