mastra-gate402

Mastra tools for Gate402 — pay-per-call GPU/CPU compute + open-model LLM inference, plus Base market data, over x402 (USDC). No vendor API keys: the agent pays the HTTP 402, and a free-tier key is auto-claimed on first use.

Install

npm install mastra-gate402

Use

import { Agent } from '@mastra/core/agent';
import { gate402Tools } from 'mastra-gate402';

export const defiAgent = new Agent({
  id: 'defi-agent',
  name: 'DeFi Agent',
  instructions: 'Vet Base tokens before acting: check risk, liquidity, and momentum.',
  model: 'openai/gpt-5.5',
  tools: { ...gate402Tools }
});

Or import individual tools:

import { gate402TokenRisk, gate402Infer } from 'mastra-gate402';

Tools

| Tool | What it does | |---|---| | gate402_compute | Rent metered GPU/CPU to run a container job, per-second; signed execution receipt | | gate402_token_risk | Rug/tradeability verdict + live honeypot sim for a Base token | | gate402_dex | Live DEX price / liquidity / 24h volume | | gate402_momentum | Momentum + order-flow signal | | gate402_best_swap | Best pool + estimated price impact for a trade size | | gate402_onchain | Wallet/token intel (balances, EOA/contract, tx count) | | gate402_launches | Freshest Base token launches | | gate402_infer | Per-token open-model LLM inference (Llama 3.1 / Qwen / mistral-small) |

Config (optional)

| Env | Default | Purpose | |---|---|---| | GATE402_API_KEY | (auto free tier) | Funded Gate402 account key to skip the free tier | | GATE402_BASE_URL | https://gate402.app | Override the gateway |

Each tool is a pay-per-call Gate402 endpoint settled over x402 (USDC on Base) — the payment is the auth. Full catalog: https://gate402.app/v1.

MIT.