AI-Powered Features
AI-powered features in CAST Imaging Express, the AI providers they run on, and how to enable them
Use of AI features is subject to the Artificial Intelligence and Machine Learning Terms.
Overview
CAST Imaging Express includes AI-powered features that help you understand complex application architectures through natural-language explanations and queries. This page lists the features, the AI providers they run on and how an administrator enables them.
Available features
- Explain View - An interactive AI assistant that explains any graph view in business-friendly language and supports follow-up questions.
- Application Summary Report - AI-generated narrative descriptions embedded in the application analysis report.
- AI Transaction Summary - Natural-language summaries of what each transaction and data call graph does. Requires an Own Provider.
- Business graph (beta) - A GraphRAG knowledge graph of your application, queried through the CAST Imaging Express MCP Server. Requires an Own Provider.
- AI Mode in Global Search (beta) - Find objects by describing them in plain English instead of building filters.
Each feature page describes exactly what data is sent to the AI provider and what is never sent.
AI providers
AI processing runs on one provider, chosen for the whole organization:
- CAST AI - An enterprise configuration of Claude Sonnet, provided by Anthropic and integrated by CAST. It is covered by your subscription and requires no setup.
- Own Provider - Your organization’s own LLM credentials. This gives you control over data residency and model selection. Supported providers are OpenAI, AWS Bedrock, Claude AI and Gemini.
Enabling AI features
Out of the box, all AI features are disabled. An organization administrator must enable them, choose a provider and select which features are available to users. See AI Settings.