Mandate Layer v1.0

Your AI agents spend money. You should control how.

QuetraAI sits between your agent and every vendor, enforcing cryptographically signed spending mandates with sub-50ms policy evaluation.

quetra evaluate

$ quetra evaluate --amount 1200 --vendor stripe.com

evaluating mandate for agent "research-bot"...

checking 4 rules: category, vendor, rate_limit, max_amount

APPROVEDin 12ms · budget remaining: $38.00 / $50.00

$ quetra evaluate --amount 15000 --vendor casino.com

REJECTED vendor "casino.com" not in allowlist

Latency
<50ms
p99 evaluation time

The Infrastructure Problem

Uncontrolled Agent Spending

Autonomous agents access paid APIs, purchase resources, and process payments without guardrails or approval workflows.

Budget Overruns

Without per-agent spending limits and rate controls, costs spiral — especially across multi-agent pipelines.

Governance Vacuum

No audit trail, no permissioning, no vendor controls. Enterprise-scale AI deployment becomes a legal and operational risk.

01
FRAGMENTATION
02
LATENCY SPIKES
03
COST OVERRUNS
04
RISK FACTOR

The QuetraAI Solution

A programmable governance layer that turns chaotic agent spending into controlled, auditable commerce.

Cryptographic Mandates

Ed25519-signed spending mandates define what each agent can spend, on which vendors, and when — tamper-proof and auditable.

Sub-50ms Enforcement

Real-time policy evaluation on Cloudflare Workers edge. Every transaction checked against budget, rate limits, and 7 rule types.

Complete Audit Trail

Append-only transaction log with approve/reject decisions, rule evaluation details, and budget snapshots for every spend event.

See It In Action

Monitor every agent transaction, manage mandates, and track budgets from a single dashboard.

QuetraAI dashboard showing active agents, mandate budgets, spending utilization, and recent transactions

How It Works

From zero to governed agent spending in under 5 minutes.

01

Register Agents & Set Mandates

Define spending rules, budget limits, allowed vendors, and rate controls for each AI agent through the dashboard or API.

02

Install SDK or MCP Server

Three lines of code. Works with any agent framework — LangChain, Claude, custom agents, or via MCP for zero-code integration.

03

Every Spend is Governed

Each transaction is evaluated against your mandate rules in real-time. Approved, rejected, or logged — all in under 50ms.

Multiple Integration Paths

Choose the approach that fits your stack — from zero-code MCP to full programmatic control.

SDK Integration

Drop-in TypeScript client that wraps fetch with automatic x402 payment negotiation, mandate evaluation, and budget tracking.

  • Automatic x402 payment flow
  • Budget-aware fetch()
  • Works with any agent framework
import { QuetraClient } from "@quetra/sdk";

const client = new QuetraClient({
  apiKey: process.env.QUETRA_API_KEY,
  agentId: "research-bot",
});

// Every fetch is governed by your mandate
const data = await client.fetch("https://api.example.com/data");

MCP Server

Zero-code governance for Claude, OpenClaw, and any MCP-compatible agent. Just add the server config.

MCP Setup Guide

REST API

59 endpoints for full programmatic control. Evaluate, manage mandates, track transactions from any language.

API Reference

Everything You Need

A complete governance platform for AI agent commerce.

7 Rule Types

Category, vendor allowlist, max amount, rate limit, time window, geo restriction, and custom rules.

Budget Periods

Daily, weekly, or monthly budgets with automatic inline reset. No cron jobs needed.

Real-time Webhooks

HMAC-SHA256 signed events for approvals, rejections, budget warnings at 80%, and exhaustion alerts.

On-Chain Enforcement

Optional Base L2 enforcement via Safe Smart Accounts with USDC allowance modules.

Multi-Protocol Support

x402 payment protocol, Stripe ACP, Stripe credit card governance — all through one mandate.

Dashboard

Register agents, compose mandates, monitor transactions, and manage API keys from a single interface.

Simple, Transparent Pricing

Start free. Scale as your agents grow.

Free

For testing and small projects

$0forever
  • 500 evaluations/month
  • 1 API key
  • Community support
  • All rule types
Get Started
Most Popular

Pro

For production agent deployments

$49/month
  • 10,000 evaluations/month
  • Unlimited API keys
  • Webhooks
  • Priority support
  • Usage analytics
Start Free Trial

Enterprise

For teams running agent fleets

$299/month
  • 100,000 evaluations/month
  • On-chain enforcement
  • SSO & audit logs
  • Dedicated support
  • Custom SLA
Contact Sales

Scale with Confidence.

Start governing your AI agent spending in 5 minutes. Free tier included. No credit card required.