nano-toolset
7 deterministic Nano payment tools for autonomous AI agents.
MCP over stdio and HTTP API both in one binary.
$ cargo run --release
HTTP API listening on 127.0.0.1:3123 MCP active on stdio
Why nano-toolset
Built to give agents a deterministic way to execute payment actions without framework lock-in.
Deterministic
Consistent success/data/error envelope on every response. Your agent always knows what to expect.
Framework-Agnostic
MCP for native agent frameworks. HTTP API for everything else. Python, JavaScript, curl, Rust any client works.
Fast & Lean
Written in Rust. Single binary, minimal resource usage, sub-second Nano transactions. No separate Python runtime or Docker requirement.
7 Payment Tools
Available via both MCP and HTTP API. Same tools, same data, your choice of transport.
wallet.balance
/wallet/balance
Retrieve current wallet balance (NANO) and pending incoming amount.
wallet.send
/wallet/send
Send Nano to any address. Sub-second finality, zero transaction fees.
payment.request
/payment/request
Create a payment request with a unique transaction_id for tracking.
payment.status
/payment/status/{'{transaction_id}'}
Poll payment status. Returns is_paid: true/false.
credits.get
/credits
Get current credit balance and all available topup price tiers.
credits.topup
/credits/topup/{'{credits_amount}'}
Top up credits. Toolset auto-sends the Nano payment.
donate.send
/donate/{'{amount}'}
Donate Nano to the platform. Supports continued development.
Architecture
One process, two transports, same wallet state.
Your Agent
MCP framework, Python, JS, curl anything
nano-toolset
Rust binary wallet Nano RPC
MCP transport
- Embedded in the same process
- Communicates via stdin/stdout
- Tool schemas exposed via
tools/list - Best for MCP-native agent frameworks
HTTP transport
- Listens on
127.0.0.1:3123 - Standard REST GET/POST
- Configurable via
HTTP_LISTEN_ADDR - Best for custom/lightweight integrations
Documentation
Everything you need to integrate.
Integration Guide
Step-by-step MCP and HTTP API setup with code examples
AI Agent Playbook
Practical agent lifecycle, payment flow patterns, and production guardrails
API Reference
Full REST API reference with request/response schemas
Code Examples
Ready-to-use Python and curl examples for all 7 tools
GitHub
Source code, issues, and contributions
Start Building with nano-toolset
Clone and run. On first start, .env is created from .env-example if missing.