Quick Start Guide

nano-toolset up and running in under 5 minutes

[AI]

nano-toolset

FOR AI AGENTS

Single Rust binary. HTTP API + MCP over stdio. Works with any agent framework or HTTP client.

  • - HTTP API on :3123
  • - MCP over stdio
  • - 7 payment tools
  • - Self-hosted
Get Started ->
[MCP]

MCP Integration

AGENT FRAMEWORKS

Use toolset as an MCP server. Works natively with LangChain, CrewAI and other MCP-compatible frameworks.

  • - MCP over stdio
  • - tools/list + tools/call
  • - Framework native
MCP Integration Guide ->
[PAY]

Payment Pages

FOR WEBSHOPS

Use POST /payment/request to create a payment request and share the returned payment details with your customer.

  • - Payment request details
  • - Address + amount flow
  • - Status polling
View API ->
[TOOLSET]

nano-toolset Quick Start

1

Clone the repository

Terminal
git clone https://github.com/ifenpay/ifenpay-nano-toolset
cd ifenpay-nano-toolset

Or download a release binary from the GitHub releases page.

2

Auto setup (no manual wallet config)

Terminal
cargo run --release
# First run creates .env (if missing) and auto-generates AI_WALLET_PASSWORD

Important wallet encryption note

AI_WALLET_PASSWORD=auto-generated on first start
HTTP_LISTEN_ADDR=127.0.0.1:3123 # optional override

Do not remove or change AI_WALLET_PASSWORD after wallet creation unless you also replace data/ai.bin. This password is bound to the wallet encryption key.

3

Start the toolset

Terminal
cargo run --release

HTTP API listening on 127.0.0.1:3123

MCP server active on stdio

Both transports start in the same process. The HTTP API and MCP share the same wallet state.

4

Make your first API call

curl
curl http://127.0.0.1:3123/wallet/balance
Python
import requests
resp = requests.get("http://127.0.0.1:3123/wallet/balance").json()
print(resp["data"]["balance"])  # "1.5" NANO

You're Ready!

Your toolset is running. All 7 payment tools are available via HTTP API and MCP.

[AGENT]

Example Agent

EXAMPLE ONLY

Disclaimer: The ifenpay-agent-example is a reference implementation for demonstration purposes only. It has no stability or compatibility guarantees. nano-toolset must be started separately before the agent.

The example agent is a Flask + Ollama application that calls nano-toolset via its HTTP API (MCP_TOOLSET_BASE_URL=http://host.docker.internal:3123).

1

Start nano-toolset

cargo run --release
2

Configure agent

cp .env-example .env
3

Start agent

docker compose up
View ifenpay-agent-example on GitHub

Integration Guide

Full HTTP API + MCP integration guide with all code examples

Read Guide ->

Community

Join our Discord for help and discussions

Join Discord ->

GitHub

Source code, issues, and open source contributions

View GitHub ->