This page provides installation and usage guide for pingpong-checkout.skill, helping you integrate PingPong Checkout payment capabilities into AI tools that support custom Skills.

When you should install the Skill

If you only want AI to read the latest documentation and API definitions, start with LLM Documentation Integration.
Install pingpong-checkout.skill when you also want the AI to review plans with PingPong Checkout-specific workflow, risk reminders, and launch checklists.

Quick Start

Three Steps to Install

1. Download: Click to download Skill package(Open in new window)
2. Import: Import pingpong-checkout.skill into your AI tool
3. Use: Start a new session and clearly state your integration requirements

Please use the pingpong-checkout skill to help review this PingPong Checkout integration plan.

Skill Features

Feature	Description
Integration Assistance	Supports both PingPong Checkout and API Only integration methods
Plan Review	Converges integration plans based on latest documentation and rules, avoiding common pitfalls
Risk Identification	Identifies key risk points like signatures, async notifications, and idempotency
Acceptance Support	Generates pre-launch verification checklists to assist with joint testing and acceptance

What the Skill does not replace

• Real integration testing for payment, notification, and refund flows
• Human code review and security review
• Production enablement, payment method activation, and acceptance testing

Important Disclaimer

All code generated by the Skill is AI-assisted and for reference only:

• ⚠️ Not Official Code: Code is not provided or maintained by PingPong officially
• ⚠️ Requires Human Review: Developers must carefully review, test, and validate all code
• ⚠️ Your Responsibility: Users are responsible for any issues arising from using generated code
• ⚠️ Follow Best Practices: Ensure code complies with your project's security standards and coding guidelines

Recommendation: Use Skill-generated code as reference templates and adapt/optimise according to your actual project requirements.

Pre-Installation Checklist

• AI tool supports custom Skill import or local directory loading
• Engineering code for integration is ready
• Sandbox merchant parameters and target payment method list obtained
• AI tool can access PingPongCheckout documentation site

Directory Structure Requirements

If manually extracting the Skill, preserve the pingpong-checkout/ directory structure. Do not modify or delete files like scripts/, references/, SKILL.md, etc.

Installation Methods

Method 1: Direct Import (Recommended)

For AI tools that support direct .skill file import.

1. Download pingpong-checkout.skill
2. Choose to import Skill in your tool
3. Select the downloaded file to complete import

Method 2: Directory Loading

For AI tools that need to load Skills from a local directory.

1. Extract pingpong-checkout.skill
2. Place the pingpong-checkout/ directory in your tool's designated Skills directory
3. Restart the tool or start a new session

Usage Guide

Basic Usage

The Skill follows this workflow:

Research → Ideate → Plan → Execute → Optimize → Review

First-Time Usage Tips

Providing the following information helps the Skill converge faster:

• Integration Method: PingPong Checkout / API Only
• Product Type: One-time payment / Auth+Capture / Tokenization
• Payment Methods: Visa, Klarna, Google Pay, etc.
• Industry Type: Physical goods / Gaming virtual / Other virtual

Best Practices

• ✅ Real Engineering Focus: Let the Skill review your actual code, not out-of-context pseudocode
• ✅ Clear Stage: Specify whether it's sandbox integration or production launch preparation
• ✅ Comprehensive Review: Have the Skill review all related logic including payment, callbacks, and refunds at once
• ✅ Conflict Confirmation: When local implementation conflicts with documentation, stop and confirm rather than proceed directly
• ⚠️ Code Review: All AI-generated code must be manually reviewed, tested, and verified
• ⚠️ Security First: Ensure generated code complies with your security standards and coding guidelines

Installation Verification

After installation, verify the following capabilities:

• Skill can identify project structure and distinguish frontend/backend responsibilities
• Skill can provide integration recommendations based on documentation and OpenAPI
• Skill can point out key risks like signatures, async notifications, idempotency
• Skill can remind compatibility and prerequisites based on payment methods

Pre-Launch Checklist

Important Notes

Installing Skill ≠ Ready for Production Launch.

1. Skill only assists in plan convergence and cannot replace real integration testing
2. All AI-generated code must be manually reviewed to ensure safety and correctness
3. All integration plans must complete comprehensive testing before release

Required Test Scenarios

Scenario Category	Test Items
Basic Flow	Normal payment, failed payment, user cancellation, timeout
Async Notifications	Signature verification, idempotency handling, duplicate notifications, out-of-order notifications
Result Processing	Result page display, backend order query fallback
Payment Methods	Real order placement, failure rollback path

Verification by Product Type

• Auth+Capture: Authorization, confirmation, cancellation boundaries
• Tokenization: Agreement setup, token payment, authorization-cancellation flow
• Refund Business: Timeliness, full refund, partial refund, exception rollback

Environment Check

• Sandbox and production domain names, keys, and script addresses configured correctly
• Production environment payment methods activated
• HTTPS, log desensitization, private key storage secured
• Async notification endpoints reachable

Common Pitfalls

Pitfall	Correct Practice
Only testing Happy Path	Test all exception branches and edge cases
Using frontend success page as proof	Use backend order queries and async notifications as standard
Ignoring notification signature verification	Must verify signatures and handle idempotency
Not confirming production environment	Confirm payment methods activated before launch
Reusing old solutions	Re-verify documentation and Skill versions
Directly using AI code	Must manually review, test, and verify all generated code

Related Documentation

• LLM Documentation Integration
• API Usage Guidelines
• Signature Rules
• Payment Result Notifications
• Transaction Idempotency

Previous pageLLM Documentation Integration