# Cassis documentation

Cassis does context maintenance for analytics agents. It keeps the context they read current, tested, and approved as schemas and definitions change. The approved ontology generates and runs the SQL behind every answer; questions, clarifications, evals, and source changes surface gaps and drift for your data team to review.

Use Cassis through the web app or fully headless. People and agents ask through the web app, Slack, or any MCP client; data teams maintain context through the web app, Git, MCP, or CLI; warehouse integrations supply the current data behind answers.

-   [See where it fits](/start/overview/): (Understand Cassis) Follow a question from any client through Cassis to your warehouse and back as an answer.
-   [Try the guided demo](/start/try/): (Start here) Find a context gap, teach Cassis once, and see the next answer improve.
-   [Ask from an MCP client](/ask/mcp/): Connect any MCP-compatible client. Claude Code, Claude Desktop, and Cursor are examples.
-   [Request access for your warehouse](https://getcassis.com/contact/): Connect Cassis to your own data and business definitions.

## Browse by job

-   [Get started](/build/): Choose how to manage the ontology, connect data, and start modeling.
-   [Ask your data](/ask/): Get answers in the web app, from any MCP client, or in Slack.
-   [Curate and govern your context](/curate/): Review issues and move app-managed or git-managed changes safely into production.
-   [Admin](/setup/): Manage projects, members, permissions, keys, and integrations.