MCPInstall

agentic-commerce-protocol

by reason-machinesAI & Agents

Integrate the Agentic Commerce Protocol (ACP) for AI-driven commerce between buyers, agents, and businesses

Installs
154
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/reason-machines/ai-agent-skills --skill agentic-commerce-protocol
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads agentic-commerce-protocol automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

Agentic Commerce Protocol (ACP) Skill

Skill by ara.so — AI Agent Skills collection.

Overview

The Agentic Commerce Protocol (ACP) is an open standard for connecting buyers, their AI agents, and businesses to complete purchases seamlessly. Maintained by OpenAI and Stripe, ACP enables AI agents to initiate and complete commerce transactions on behalf of users.

Key Capabilities:

  • AI agents can discover and purchase products/services
  • Businesses expose checkout endpoints to AI agents
  • Payment providers handle secure payment token exchange
  • Support for carts, orders, fulfillment, and extensions
  • Date-based versioning (YYYY-MM-DD format)

Current Version: 2026-04-17

Installation & Setup

For Merchants (Implementing Checkout Endpoints)

  1. Clone the specification repository:
git clone https://github.com/agentic-commerce-protocol/agentic-commerce-protocol.git
cd agentic-commerce-protocol
  1. Review the OpenAPI specification:
# Latest stable spec
cat spec/2026-04-17/openapi/openapi.agentic_checkout.yaml
  1. Install dependencies for your implementation:

Node.js/Express example:

npm init -y
npm install express body-parser express-validator

Python/Flask example:

pip install flask flask-cors jsonschema

For AI Agent Developers

Review the specification and integrate with a reference implementation:

Core Concepts

1. Agentic Checkout Flow

The basic checkout flow involves these endpoints:

  • POST /agentic-checkout/capabilities - Discover seller capabilities
  • POST /agentic-checkout/create-checkout - Initialize checkout session
  • POST /agentic-checkout/update-checkout - Update checkout details
  • POST /agentic-checkout/confirm-checkout - Finalize and confirm purchase

2. Payment Handlers

ACP supports multiple payment handler types:

  • Delegate Payment Handler - Agent handles payment collection
  • Seller-Backed Payment Handler - Seller collects payment directly
  • Payment Link Handler - Redirect to payment page

Implementation Examples

Basic Express.js Seller Implementation

const express = require('express');
const bodyParser = require('body-parser');
const { v4: uuidv4 } = require('uuid');

const app = express();
app.use(bodyParser.json());

// In-memory store (use database in production)
const checkoutSessions = new Map();

// Capabilities endpoint
app.post('/agentic-checkout/capabilities', (req, res) => {
  res.json({
    version: '2026-04-17',
    supported_features: [
      'cart',
      'orders',
      'fulfillment',
      'extensions'
    ],
    payment_handlers: [
      {
        type: 'delegate_payment',
        supported_methods: ['card', 'bank_transfer']
      }
    ],
    extensions: [
      {
        type: 'discount',
        supported_discount_types: ['percentage', 'fixed_amount']
      }
    ]
  });
});

// Create checkout endpoint
app.post('/agentic-checkout/create-checkout', (req, res) => {
  const { items, buyer_info, payment_handler } = req.body;
  
  // Validate required fields
  if (!items || !Array.isArray(items) || items.length === 0) {
    return res.status(400).json({
      error: {
        type: 'invalid_request',
        message: 'Items are required'
      }
    });
  }

  // Create session
  const sessionId = uuidv4();
  const session = {
    id: sessionId,
    status: 'open',
    items: items.map(item => ({
      id: item.id,
      name: item.name,
      quantity: item.quantity || 1,
      unit_price: item.unit_price,
      total: (item.quantity || 1) * item.unit_price
    })),
    buyer_info: buyer_info || {},
    payment_handler,
    created_at: new Date().toISOString()
  };

  // Calculate totals
  session.subtotal = session.items.reduce((sum, item) => sum + item.total, 0);
  session.total = session.subtotal;

  checkoutSessions.set(sessionId, session);

  res.json({
    checkout_id: sessionId,
    status: 'open',
    items: session.items,
    subtotal: session.subtotal,
    total: session.total,
    payment_handler: session.payment_handler
  });
});

// Update checkout endpoint
app.post('/agentic-checkout/update-checkout', (req, res) => {
  const { checkout_id, items, buyer_info, shipping_address } = req.body;

  const session = checkoutSessions.get(checkout_id);
  if (!session) {
    return res.status(404).json({
      error: {
        type: 'checkout_not_found',
        message: 'Checkout session not found'
      }
    });
  }

  if (session.status !== 'open') {
    return res.status(400).json({
      error: {
        type: 'invalid_state',
        message: 'Checkout is not in open state'
      }
    });
  }

  // Update session
  if (items) {
    session.items = items.map(item => ({
      id: item.id,
      name: item.name,
      quantity: item.quantity || 1,
      unit_price: item.unit_price,
      total: (item.quantity || 1) * item.unit_price
    }));
    session.subtotal = session.items.reduce((sum, item) => sum + item.total, 0);
    session.total = session.subtotal;
  }

  if (buyer_info) {
    session.buyer_info = { ...session.buyer_info, ...buyer_info };
  }

  if (shipping_address) {
    session.shipping_address = shipping_address;
  }

  res.json({
    checkout_id: session.id,
    status: session.status,
    items: session.items,
    subtotal: session.subtotal,
    total: session.total
  });
});

// Confirm checkout endpoint
app.post('/agentic-checkout/confirm-checkout', (req, res) => {
  const { checkout_id, payment_token } = req.body;

  const session = checkoutSessions.get(checkout_id);
  if (!session) {
    return res.status(404).json({
      error: {
        type: 'checkout_not_found',
        message: 'Checkout session not found'
      }
    });
  }

  if (session.status !== 'open') {
    return res.status(400).json({
      error: {
        type: 'invalid_state',
        message: 'Checkout is not in open state'
      }
    });
  }

  // Process payment (integrate with payment processor)
  // For demo purposes, we'll simulate success
  session.status = 'completed';
  session.payment_token = payment_token;
  session.completed_at = new Date().toISOString();

  const orderId = `ord_${uuidv4()}`;
  session.order_id = orderId;

  res.json({
    checkout_id: session.id,
    status: 'completed',
    order_id: orderId,
    confirmation: {
      order_number: orderId,
      total: session.total,
      items: session.items
    }
  });
});

app.listen(3000, () => {
  console.log('ACP server listening on port 3000');
});

Python/Flask Seller Implementation

from flask import Flask, request, jsonify
from flask_cors import CORS
import uuid
from datetime import datetime

app = Flask(__name__)
CORS(app)

# In-memory store
checkout_sessions = {}

@app.route('/agentic-checkout/capabilities', methods=['POST'])
def capabilities():
    return jsonify({
        'version': '2026-04-17',
        'supported_features': [
            'cart',
            'orders',
            'fulfillment'
        ],
        'payment_handlers': [
            {
                'type': 'delegate_payment',
                'supported_methods': ['card']
            }
        ]
    })

@app.route('/agentic-checkout/create-checkout', methods=['POST'])
def create_checkout():
    data = request.json
    items = data.get('items', [])
    
    if not items:
        return jsonify({
            'error': {
                'type': 'invalid_request',
                'message': 'Items are required'
            }
        }), 400
    
    session_id = str(uuid.uuid4())
    
    # Process items
    processed_items = []
    subtotal = 0
    
    for item in items:
        quantity = item.get('quantity', 1)
        unit_price = item.get('unit_price', 0)
        total = quantity * unit_price
        
        processed_items.append({
            'id': item.get('id'),
            'name': item.get('name'),
            'quantity': quantity,
            'unit_price': unit_price,
            'total': total
        })
        subtotal += total
    
    session = {
        'id': session_id,
        'status': 'open',
        'items': processed_items,
        'buyer_info': data.get('buyer_info', {}),
        'payment_handler': data.get('payment_handler'),
        'subtotal': subtotal,
        'total': subtotal,
        'created_at': datetime.utcnow().isoformat()
    }
    
    checkout_sessions[session_id] = session
    
    return jsonify({
        'checkout_id': session_id,
        'status': 'open',
        'items': processed_items,
        'subtotal': subtotal,
        'total': subtotal
    })

@app.route('/agentic-checkout/confirm-checkout', methods=['POST'])
def confirm_checkout():
    data = request.json
    checkout_id = data.get('checkout_id')
    payment_token = data.get('payment_token')
    
    session = checkout_sessions.get(checkout_id)
    if not session:
        return jsonify({
            'error': {
                'type': 'checkout_not_found',
                'message': 'Checkout session not found'
            }
        }), 404
    
    if session['status'] != 'open':
        return jsonify({
            'error': {
                'type': 'invalid_state',
                'message': 'Checkout is not in open state'
            }
        }), 400
    
    # Process payment
    session['status'] = 'completed'
    session['payment_token'] = payment_token
    order_id = f"ord_{uuid.uuid4()}"
    session['order_id'] = order_id
    
    return jsonify({
        'checkout_id': checkout_id,
        'status': 'completed',
        'order_id': order_id,
        'confirmation': {
            'order_number': order_id,
            'total': session['total'],
            'items': session['items']
        }
    })

if __name__ == '__main__':
    app.run(port=3000, debug=True)

Capability Negotiation

// Client-side capability check
async function checkSellerCapabilities(sellerUrl) {
  const response = await fetch(`${sellerUrl}/agentic-checkout/capabilities`, {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
    }
  });
  
  const capabilities = await response.json();
  
  // Check if seller supports required features
  const requiredFeatures = ['cart', 'orders'];
  const hasRequiredFeatures = requiredFeatures.every(feature =>
    capabilities.supported_features?.includes(feature)
  );
  
  return {
    compatible: hasRequiredFeatures,
    capabilities
  };
}

Extension Support (Discounts)

// Apply discount extension
app.post('/agentic-checkout/update-checkout', (req, res) => {
  const { checkout_id, extensions } = req.body;
  const session = checkoutSessions.get(checkout_id);
  
  if (!session) {
    return res.status(404).json({ error: 'Session not found' });
  }
  
  // Process discount extension
  const discountExt = extensions?.find(ext => ext.type === 'discount');
  if (discountExt) {
    const { discount_type, value, code } = discountExt;
    
    // Validate discount code
    if (code === 'SAVE10') {
      if (discount_type === 'percentage') {
        session.discount = session.subtotal * (value / 100);
      } else if (discount_type === 'fixed_amount') {
        session.discount = value;
      }
      
      session.total = session.subtotal - session.discount;
    }
  }
  
  res.json({
    checkout_id: session.id,
    subtotal: session.subtotal,
    discount: session.discount || 0,
    total: session.total
  });
});

Order Management

// Get order status
app.post('/agentic-checkout/get-order', (req, res) => {
  const { order_id } = req.body;
  
  // Find session by order_id
  const session = Array.from(checkoutSessions.values())
    .find(s => s.order_id === order_id);
  
  if (!session) {
    return res.status(404).json({
      error: {
        type: 'order_not_found',
        message: 'Order not found'
      }
    });
  }
  
  res.json({
    order_id: session.order_id,
    status: session.status,
    created_at: session.created_at,
    items: session.items,
    total: session.total,
    fulfillment: {
      status: 'pending',
      tracking_number: null
    }
  });
});

Configuration

Environment Variables

# Server configuration
PORT=3000
NODE_ENV=production

# Payment processor
PAYMENT_PROCESSOR_API_KEY=your_payment_api_key
PAYMENT_PROCESSOR_SECRET=your_payment_secret

# Security
WEBHOOK_SECRET=your_webhook_secret
ALLOWED_ORIGINS=https://agent1.example.com,https://agent2.example.com

# Database
DATABASE_URL=postgresql://user:password@localhost:5432/acp_db

# ACP version
ACP_VERSION=2026-04-17

Configuration File Example

// config/acp.config.js
module.exports = {
  version: process.env.ACP_VERSION || '2026-04-17',
  
  features: {
    cart: true,
    orders: true,
    fulfillment: true,
    extensions: true,
    authentication: false
  },
  
  paymentHandlers: [
    {
      type: 'delegate_payment',
      methods: ['card', 'bank_transfer'],
      processor: 'stripe'
    }
  ],
  
  extensions: [
    {
      type: 'discount',
      supportedTypes: ['percentage', 'fixed_amount', 'code']
    }
  ],
  
  security: {
    requireAuth: false,
    allowedOrigins: process.env.ALLOWED_ORIGINS?.split(',') || [],
    webhookSecret: process.env.WEBHOOK_SECRET
  }
};

Common Patterns

1. Idempotent Request Handling

const processedRequests = new Map();

app.post('/agentic-checkout/confirm-checkout', (req, res) => {
  const idempotencyKey = req.headers['idempotency-key'];
  
  if (idempotencyKey && processedRequests.has(idempotencyKey)) {
    // Return cached response
    return res.json(processedRequests.get(idempotencyKey));
  }
  
  // Process request
  const result = processCheckout(req.body);
  
  if (idempotencyKey) {
    processedRequests.set(idempotencyKey, result);
  }
  
  res.json(result);
});

2. Webhook Event Handling

app.post('/webhooks/acp', (req, res) => {
  const signature = req.headers['x-acp-signature'];
  
  // Verify signature
  if (!verifyWebhookSignature(req.body, signature)) {
    return res.status(401).json({ error: 'Invalid signature' });
  }
  
  const event = req.body;
  
  switch (event.type) {
    case 'checkout.completed':
      handleCheckoutCompleted(event.data);
      break;
    case 'checkout.cancelled':
      handleCheckoutCancelled(event.data);
      break;
    case 'order.fulfilled':
      handleOrderFulfilled(event.data);
      break;
  }
  
  res.json({ received: true });
});

3. Error Handling

// Standard error response format
function sendError(res, type, message, statusCode = 400) {
  res.status(statusCode).json({
    error: {
      type,
      message,
      timestamp: new Date().toISOString()
    }
  });
}

// Usage
app.post('/agentic-checkout/create-checkout', (req, res) => {
  try {
    const { items } = req.body;
    
    if (!items || items.length === 0) {
      return sendError(res, 'invalid_request', 'Items are required');
    }
    
    // Process checkout...
    
  } catch (error) {
    console.error('Checkout error:', error);
    sendError(res, 'internal_error', 'An error occurred', 500);
  }
});

4. Cart Management

// Add item to cart
app.post('/agentic-checkout/add-to-cart', (req, res) => {
  const { checkout_id, item } = req.body;
  const session = checkoutSessions.get(checkout_id);
  
  if (!session) {
    return sendError(res, 'checkout_not_found', 'Session not found', 404);
  }
  
  // Add or update item
  const existingIndex = session.items.findIndex(i => i.id === item.id);
  
  if (existingIndex >= 0) {
    session.items[existingIndex].quantity += item.quantity || 1;
  } else {
    session.items.push({
      id: item.id,
      name: item.name,
      quantity: item.quantity || 1,
      unit_price: item.unit_price,
      total: (item.quantity || 1) * item.unit_price
    });
  }
  
  // Recalculate totals
  session.subtotal = session.items.reduce((sum, i) => sum + i.total, 0);
  session.total = session.subtotal - (session.discount || 0);
  
  res.json({
    checkout_id: session.id,
    items: session.items,
    subtotal: session.subtotal,
    total: session.total
  });
});

Testing

Test Suite Example

const request = require('supertest');
const app = require('./server');

describe('ACP Checkout Flow', () => {
  let checkoutId;
  
  test('GET capabilities', async () => {
    const res = await request(app)
      .post('/agentic-checkout/capabilities')
      .expect(200);
    
    expect(res.body.version).toBe('2026-04-17');
    expect(res.body.supported_features).toContain('cart');
  });
  
  test('POST create-checkout', async () => {
    const res = await request(app)
      .post('/agentic-checkout/create-checkout')
      .send({
        items: [
          {
            id: 'prod_123',
            name: 'Test Product',
            quantity: 2,
            unit_price: 1000
          }
        ],
        payment_handler: {
          type: 'delegate_payment'
        }
      })
      .expect(200);
    
    checkoutId = res.body.checkout_id;
    expect(res.body.status).toBe('open');
    expect(res.body.total).toBe(2000);
  });
  
  test('POST confirm-checkout', async () => {
    const res = await request(app)
      .post('/agentic-checkout/confirm-checkout')
      .send({
        checkout_id: checkoutId,
        payment_token: 'tok_test_12345'
      })
      .expect(200);
    
    expect(res.body.status).toBe('completed');
    expect(res.body.order_id).toBeDefined();
  });
});

Troubleshooting

Common Issues

1. Invalid version error

Error: Unsupported ACP version

Solution: Ensure your implementation supports the version specified in the request. Check spec/2026-04-17/ for the current stable version.

2. Payment handler mismatch

Error: Unsupported payment handler type

Solution: Verify the payment handler type in your capabilities response matches what the agent is requesting.

3. Checkout session not found

Error: checkout_not_found

Solution: Implement proper session storage (Redis, database) instead of in-memory maps for production. Sessions should persist across server restarts.

4. CORS errors

Access-Control-Allow-Origin header missing

Solution: Add CORS middleware with appropriate allowed origins:

const cors = require('cors');
app.use(cors({
  origin: process.env.ALLOWED_ORIGINS.split(','),
  credentials: true
}));

5. Idempotency issues

Duplicate checkout created

Solution: Implement idempotency key handling for all mutating operations.

Validation

Use the JSON schemas provided in the spec for validation:

const Ajv = require('ajv');
const ajv = new Ajv();

const checkoutSchema = require('./spec/2026-04-17/json-schema/checkout.schema.json');
const validate = ajv.compile(checkoutSchema);

app.post('/agentic-checkout/create-checkout', (req, res) => {
  const valid = validate(req.body);
  if (!valid) {
    return res.status(400).json({
      error: {
        type: 'validation_error',
        message: validate.errors
      }
    });
  }
  // Process request...
});

Additional Resources

  • Official Documentation: https://agenticcommerce.dev
  • OpenAPI Spec: spec/2026-04-17/openapi/openapi.agentic_checkout.yaml
  • Examples: examples/2026-04-17/
  • Governance: docs/governance.md
  • Contributing: CONTRIBUTING.md

FAQ

What does the agentic-commerce-protocol skill do?

Integrate the Agentic Commerce Protocol (ACP) for AI-driven commerce between buyers, agents, and businesses

How do I install agentic-commerce-protocol?

Run npx skills add https://github.com/reason-machines/ai-agent-skills --skill agentic-commerce-protocol in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/agentic-commerce-protocol/ (personal) or .claude/skills/agentic-commerce-protocol/ (project).

Which AI agents can use agentic-commerce-protocol?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is agentic-commerce-protocol safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains agentic-commerce-protocol?

It is published by reason-machines in reason-machines/ai-agent-skills. It has 154 recorded installs.

More from reason-machines/ai-agent-skills

View all →

agency-agents-zh-skill

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents322 installs

agent-reach-internet-access

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents228 installs

agent-sprite-forge-2d-game-assets

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

Media & Creative216 installs

agentmemory-persistent-memory

reason-machines/ai-agent-skills

Add persistent memory to AI coding agents using agentmemory - remembers context, preferences, and decisions across sessions

AI & Agents216 installs

awesome-agent-skills

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents214 installs

oh-my-openagent-orchestration

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents211 installs

Popular ai & agents skills

View category →

google-agents-cli-publish

google/agents-cli

This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).

AI & Agents513.9K installs

google-agents-cli-eval

google/agents-cli

This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).

AI & Agents513.5K installs

skill-creator

anthropics/skills

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

OfficialAI & Agents416.4K installs

codex-pet

prime-skills/runcomfy-agent-skills

Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.

AI & Agents402.1K installs

full-output-enforcement

leonxlnx/taste-skill

Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.

AI & Agents401K installs

executing-plans

obra/superpowers

Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available

AI & Agents245.7K installs