Skip to main content

​
Overview

ACN’s marketplace is organized into two levels:
  • Providers — Companies or services that offer capabilities (e.g., Resend, CoinGecko, Alchemy)
  • Endpoints — Specific actions a provider offers (e.g., “send email”, “get price”, “get block”)
A single provider can have multiple endpoints. Each endpoint has its own pricing, schema, and quality metrics.

​
Provider Profile

Every provider on ACN has a public profile: 
{
  "provider_id": "resend",
  "name": "Resend",
  "description": "Modern email API for developers",
  "category": "communication",
  "website": "https://resend.com",
  "endpoints": [
    {
      "slug": "send-email",
      "description": "Send transactional emails",
      "pricing": { "model": "per_call", "price_per_call_usdc": "0.002" }
    },
    {
      "slug": "send-batch",
      "description": "Send batch emails (up to 100 recipients)",
      "pricing": { "model": "per_call", "price_per_call_usdc": "0.01" }
    }
  ]
}

​
Service Categories

Providers are organized into categories:
CategoryExamples
Data IntelligenceWeb search, financial data, news, weather
AI / MLText generation, image generation, speech-to-text, embeddings
CommunicationEmail, SMS, push notifications, messaging
BlockchainOn-chain data, token prices, transaction history, oracles
UtilityGeocoding, document parsing, file conversion, scheduling

​
Endpoint Schema

Each endpoint defines a strict schema for requests and responses: 
{
  "slug": "send-email",
  "method": "POST",
  "description": "Send a transactional email with optional HTML body",
  "parameters": {
    "to": { "type": "string", "required": true, "description": "Recipient email" },
    "subject": { "type": "string", "required": true, "description": "Email subject" },
    "body": { "type": "string", "required": false, "description": "Plain text body" },
    "html": { "type": "string", "required": false, "description": "HTML body" }
  },
  "response_schema": {
    "message_id": { "type": "string" },
    "status": { "type": "string", "enum": ["sent", "queued"] }
  }
}
ACN validates requests against this schema before proxying to the provider. This means agents get clear, immediate errors for malformed requests rather than confusing provider-specific error messages.

​
Quality Metrics

Every endpoint has live quality metrics, updated continuously:
MetricDescription
uptime_pctPercentage of successful responses (non-5xx) over the trailing period
avg_latency_msAverage response time in milliseconds
error_rate_pctPercentage of calls resulting in errors
total_callsTotal number of calls served (lifetime)
These metrics are included in discovery results and influence ranking. Providers with higher reliability and lower latency rank higher for equivalent relevance.

​
Pricing

Providers set their own per-call pricing in USDC: 
{
  "pricing": {
    "model": "per_call",
    "price_per_call_usdc": "0.002"
  }
}
Pricing is transparent — agents see the exact cost before executing a call. There are no hidden fees at the endpoint level. See Platform Pricing for details on platform fees.

​
How Agents Use This

When an agent discovers services, all of the above — provider info, endpoint schemas, pricing, and quality metrics — is returned in the response. This gives agents everything they need to:
  1. Choose the best provider based on quality, price, and relevance
  2. Construct a valid request using the endpoint schema
  3. Estimate the cost before executing
  4. Evaluate reliability based on quality metrics