多平台 AI 助手插件快速开发方案设计器

You are a senior AI platform engineer specializing in building plugins and extensions for AI assistant platforms. Design a complete plugin development plan based on the requirements below.

Plugin Requirements

Target platforms: [e.g., OpenAI GPTs / Claude MCP Server / Gemini Extensions / All] Plugin functionality: [Describe what the plugin should do] Data sources: [e.g., REST API / Database / File system / External service] Auth requirements: [e.g., OAuth2 / API key / No auth] Deployment target: [e.g., Cloudflare Workers / AWS Lambda / Self-hosted]

Output Structure

1. Architecture Overview

• System diagram (Mermaid format)
• Data flow between AI platform - Plugin - Backend
• Security boundary analysis

2. API Specification

• OpenAPI 3.1 schema for all endpoints
• Request/response examples
• Error handling patterns
• Rate limiting strategy

3. Platform-Specific Configurations

For each target platform, provide:

• Manifest/config file (e.g., ai-plugin.json, MCP server config)
• Tool/function definitions
• Permission scopes needed
• Platform-specific limitations and workarounds

4. Authentication & Security

• Auth flow diagram
• Token management strategy
• Input validation rules
• Data privacy considerations (PII handling)

5. Development Roadmap

• Phase 1: MVP (core functionality)
• Phase 2: Enhanced features
• Phase 3: Multi-platform deployment
• Estimated timeline for each phase

6. Testing Strategy

• Unit test approach for plugin logic
• Integration test with AI platform sandbox
• Load testing plan
• Security audit checklist

7. Deployment & Monitoring

• CI/CD pipeline configuration
• Health check endpoints
• Logging and observability setup
• Rollback strategy

Provide all code examples in the most appropriate language for the deployment target. Include ready-to-use boilerplate code.