/
Supportvalsight.ai
Open in ClaudeOpen in ChatGPT

Valsight AI

The Valsight AI Assistant is a conversational interface that helps you understand, explore, and navigate your Valsight models using natural language. It provides contextual answers based on your model structure, formulas, metadata, and project configuration, without modifying any data.

Currently, the AI Assistant is designed primarily to explain model logic. It is available in read-only mode.

Start here if you want to:

  • Ask questions about your model without manually inspecting nodes or formulas
  • Get explanations of model structure, relationships, or calculations
  • Find out what information the AI can access and how answers are generated
  • Understand the scope, limits, and reliability of AI-generated guidance

Not here if you want to:

  • Find step-by-step modeling instructions
  • Change data, formulas, or scenarios using AI
  • Find detailed UI walkthroughs for non-AI features

Who this section is for

This section is for Valsight users who work with models and want faster insight, clarification, or orientation through natural language. It is especially useful when reviewing existing models, onboarding into complex projects, or validating understanding of calculations and structure.

How this fits into Valsight

The AI Assistant complements existing documentation, modeling tools, and UI features by providing on-demand explanations and guidance. It does not replace formal modeling workflows or documentation, but acts as a contextual support layer while you work.

The pages below cover the core concepts and behavior of the Valsight AI Assistant:

Valsight AI lets you understand, explore, and work with your Valsight models using natural language. There are two ways to use it, both built on the same model context — your model structure, formulas, metadata, and project configuration.

The two ways to use Valsight AI

  • In-app Assistant — communicate with the AI directly inside Valsight. You chat with it in the app while you work, and it answers using the context of your models. It is read-only today; the ability to have it build and change models for you is coming in a future release.
  • Valsight MCP integration — connect your own external AI client (such as Claude) to Valsight through the MCP server. The client then explores and changes your models from outside the app, inside a session whose scope and permissions you control. This is a premium feature, available from Valsight 7.3.0 and included at no additional cost through the end of 2026.

Which one to choose

  • Use the in-app Assistant when you want help without leaving Valsight — quick questions, orientation, and explanations while you work.
  • Use the Valsight MCP integration when you want to drive Valsight from your own AI client or agent, or combine it with other tools. This is the more technical option and is aimed at technical users.

The pages in this section

Was this page helpful?