Welcome to the ISC-CodeConnect documentation. This comprehensive guide will help you understand, use, and contribute to the multi-agent RAG Salesforce code assistant.
ISC-CodeConnect is an advanced AI-powered code assistant specifically designed for Salesforce development. It features:
- 🤖 Multi-Agent Architecture: Specialized agents working in orchestrated pipelines
- 🗄️ IBM CIO Milvus Integration: Production-grade vector database with connection pooling
- ⚡ Intelligent Code Generation: Apex classes, test methods, and Lightning Web Components
- 🔍 Context-Aware Retrieval: Sophisticated code search across 5 specialized collections
- 🔄 Session-Based API: FastAPI with streaming responses and state management
- 📊 Advanced Analytics: Query optimization and performance monitoring
🏗️ Architecture
- System Overview - High-level system architecture and data flow
🤖 Agents
- Agent Overview - Complete guide to all agent types and interactions
- Core Orchestration Agents:
- Supervisor Agent - Query decomposition and stage orchestration
- Router Agent - Query analysis and agent selection
- Follow-up Handler - Multi-turn conversation enhancement
- Query Analysis Agent - Query intent analysis and classification
- Research Agent - Detailed analysis and planning
- Generation Pipelines:
- Test Generation Pipeline - 5-stage Apex test generation workflow
- LWC Generation Pipeline - 4-stage Lightning Web Component generation
- Support Agents:
- Specialized Agents - Context analyzers, error handlers, and system support
- REST API Endpoints - FastAPI endpoints with session-based authentication
- Getting Started - Quick start guide for new users
- Development Setup - Local development environment configuration
- Milvus Setup - IBM CIO hosted Milvus configuration and local development
- Milvus Connection Manager - Connection pooling and health monitoring
- Retriever Factory - Vector database abstraction and specialized retrievers
- Start with Getting Started
- Review System Overview
- Follow Development Setup
- Understand Agent Overview and Agent Orchestration
- Learn about Vector Database and Retriever Factory
- Check API Reference for integration details
- Review System Overview for high-level architecture
- Understand Agent Pipelines and LWC Generation
- Study Milvus Connection Management for production deployment
- Apex Test Generation: 5-stage pipeline for comprehensive test coverage
- LWC Component Creation: 4-stage workflow for complete Lightning Web Components
- Best Practice Integration: Automatic adherence to Salesforce coding standards
- Context-Aware Search: Finds relevant code patterns across repositories
- Multi-Collection Queries: Searches Apex, LWC, and test code simultaneously
- Dependency Mapping: Automatically identifies code relationships
- Query Decomposition: Breaks complex requests into manageable stages
- Intelligent Routing: Directs queries to the most appropriate agents
- Follow-up Processing: Handles multi-turn conversations with context retention
This documentation is actively maintained and reflects the current system implementation.
- Latest Version: v1.2.0
- Documentation Version: v1.2.0
- Last Review: July 3, 2025
- Agent System: All agent types documented with detailed workflows
- Vector Database: IBM CIO Milvus integration with connection pooling
- API Endpoints: Session-based FastAPI implementation
- Generation Pipelines: Test generation and LWC generation workflows
- Development Setup: Local environment configuration
- Advanced deployment configurations
- Performance optimization guides
- Integration examples and tutorials
Found an issue or want to improve the documentation?
- Report Issues: Create an issue in the main repository
- Suggest Improvements: Submit a pull request with specific changes
- Add Examples: Contribute practical examples and use cases
- Update Accuracy: Help keep documentation aligned with code changes
- GitHub Issues: Technical questions and bug reports
- Documentation Issues: Specific documentation feedback
- Enterprise Support: Contact isc-codeconnect@ibm.com
📖 Happy Reading!