TonQuant

TON DeFi market research and trading CLI — built for AI Agents.

TON AI Agent Hackathon · Track 1: Agent Infrastructure

Website: tonquant.com · npm: npm install -g tonquant

What is TonQuant?

A command-line tool that lets AI Agents (like OpenClaw) and developers perform quantitative research, backtest strategies, and inspect DeFi markets on the TON blockchain.

Key features:

• Agent-driven research loop: autoresearch run chains data→factor→backtest→report automatically
• Full quant pipeline: OHLCV data, RSI/MACD/volatility factors, momentum backtesting
• --json output for AI agent consumption
• Human-readable colored terminal output for developers
• STON.fi DEX integration (prices, pools, swap simulation)
• TonAPI integration (wallet balance, transaction history)
• Encrypted mnemonic storage (AES-256-GCM)

Quick Start

# Install (requires Bun on PATH — https://bun.sh)
npm install -g tonquant

# Try it out
tonquant trending --limit 5
tonquant price TON
tonquant pools USDT/TON
tonquant research NOT --json

tonquant ships with its bundled quant backend in the published package. Support commands and quant commands work after install without manually setting TONQUANT_QUANT_CLI.

Commands

Quant Research (Phase 1)

Command	Description
tonquant autoresearch run --asset TON/USDT	Full research loop: data→factor→backtest→report
tonquant data fetch TON/USDT	Fetch OHLCV market data
tonquant data list	List cached datasets
tonquant factor compute --factors rsi,macd,volatility	Compute technical factors
tonquant factor list	List available factors
tonquant backtest run --strategy momentum	Run strategy backtest
tonquant preset list	List strategy presets
tonquant preset show momentum-ton	Show preset details

Market Inspection (Phase 0)

Command	Description
tonquant price <symbol>	Token price and details
tonquant pools <A>/<B>	Pool liquidity and fees
tonquant trending	Top tokens by liquidity
tonquant research <symbol>	Market research summary
tonquant init --mnemonic '...'	Configure wallet
tonquant balance [--all]	Wallet balance with USD
tonquant swap <from> <to> <amount>	Simulate DEX swap
tonquant history	Transaction history

Add --json to any command for structured JSON output.

See skill/SKILL.md for full command reference, JSON schemas, and Agent workflows.

Architecture

CLI Layer (commander)
  Phase 0: price | pools | trending | init | balance | swap | research | history
  Phase 1: data | factor | backtest | preset | autoresearch
    ↓
Quant API Layer (src/quant/api/)              Query Layer (queries.ts)
  runDataFetch | runFactorCompute |            fetchPriceData | fetchPoolData | ...
  runBacktest | runPresetShow                    ↓
    ↓                                         Service Layer
Quant Runner (src/quant/runner/)               stonfi.ts | tonapi.ts | wallet.ts
  subprocess spawn → JSON stdio                 ↓
    ↓                                         Cache Layer (cache.ts)
Quant Backend (quant-backend/)                   ↓
  data | factor | backtest | preset           External APIs
    ↓                                           STON.fi v1 | TonAPI v2
Orchestrator (src/quant/orchestrator.ts)
  preset → data → factor → backtest → report

Tech Stack

Component	Technology
Runtime	Bun
Language	TypeScript (strict)
CLI	Commander
Validation	Zod
TON SDK	@ton/ton, @ton/crypto, @ton/core
DEX	STON.fi HTTP API v1
Wallet API	TonAPI v2
Terminal UI	chalk, cli-table3
Linting	Biome
Testing	bun:test (135 tests, 80%+ coverage)

For AI Agents

TonQuant is designed as an OpenClaw skill. The skill/SKILL.md file teaches any MCP-compatible Agent how to:

1. Discover tokens (trending)
2. Research before trading (research, price, pools)
3. Simulate trades (swap)
4. Monitor portfolio (balance --all)

All commands support --json for structured parsing.

Development

bun test              # Run tests
bun test --coverage   # With coverage report
bun run typecheck     # TypeScript check
bun run lint          # Biome linting
bun run dev           # Run CLI in dev mode

License

MIT