> ## Documentation Index
> Fetch the complete documentation index at: https://docs.credibledata.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Credible App

> Chat with data, create reports, explore models, and view notebooks in workspaces

The Credible App (`https://<your-org>.app.credibledata.com`) is where your team consumes published semantic models. Everything happens inside **workspaces** — shared spaces scoped to a set of packages and members.

<img src="https://mintcdn.com/ms2/hbN0V7V2tgDWdpis/images/quickstart/Workspace-landing-page.png?fit=max&auto=format&n=hbN0V7V2tgDWdpis&q=85&s=f7aef9f2d7d0bfb2d9b9c37a74a061ff" alt="Workspace landing page showing chat input, tabs for Chats, Reports, and Models" style={{maxWidth: "100%", height: "auto"}} width="2140" height="1304" data-path="images/quickstart/Workspace-landing-page.png" />

## Chat

Type a question in the chat bar at the top of any workspace. The [Credible Context Engine](/concepts/context-engine) parses your question into phrases, matches each phrase to governed entities (dimensions, measures, views) in your semantic model using `#(doc)` and `#(index)` annotations, and the agent executes a trustworthy Malloy query against your data.

<img src="https://mintcdn.com/ms2/hbN0V7V2tgDWdpis/images/quickstart/get-context-example.png?fit=max&auto=format&n=hbN0V7V2tgDWdpis&q=85&s=7055fe2f91284a625b5a638fb8769067" alt="get_context tool showing phrase parsing and entity matching" style={{maxWidth: "100%", height: "auto"}} width="1508" height="1154" data-path="images/quickstart/get-context-example.png" />

Each chat is saved in the workspace and visible to all members. Use the **Chats** tab to find previous conversations.

### How it works

1. **`get_context`** — Parses your question into semantic phrases and matches them to data entities declared in your published models. Match quality depends on the `#(doc)` descriptions and `#(index)` annotations in your model.
2. **`execute_query`** — The agent constructs and runs a Malloy query using the matched entities, returning results and visualizations.

This is what makes answers trustworthy — the agent is grounded in governed definitions, not guessing column names or interpreting ambiguous schemas.

<Note>
  Published models must be [indexed](/how-to/modeling/publishing#indexing) before chat works. Check for a green "Indexed" tag on the package page.
</Note>

## Reports

There are two kinds of reports in a workspace:

* **Governed reports** — `.malloynb` notebooks published with a package. These are versioned, reviewed, and part of the governed model. They appear under the **Reports** tab automatically.
* **Ad-hoc reports** — Created from a chat conversation. These are based on the governed model but are not part of the published package. Save any chat analysis as a report to share it with workspace members.

See [Analyzing with Notebooks](https://docs.malloydata.dev/documentation/user_guides/notebooks) for details on creating `.malloynb` notebooks.

## Explorer

The **Models** tab lets you browse semantic models available in the workspace. Click on a model to open it in the Explorer — a visual, no-code query builder where you can select dimensions, measures, and views, build queries, and visualize results as charts and tables.

This is useful for understanding what data is available before starting a chat, or for hands-on exploration of your models. See the [Explorer documentation](https://docs.malloydata.dev/documentation/user_guides/publishing/explorer) for a full guide to the interface.

## Notebooks

[Malloy notebooks](https://docs.malloydata.dev/documentation/user_guides/notebooks) (`.malloynb` files) published with a package appear as reports in the workspace. They blend markdown, Malloy queries, and visualizations into interactive documents — useful for data stories, KPI dashboards, and package overviews.

A `README.malloynb` at the package root automatically renders on the package landing page.

## Getting Started

1. **Publish and index a model** — See [Publishing](/how-to/modeling/publishing)
2. **Create a workspace** — Add packages and members in the [Credible App](/platform-admin/admin-portal#creating-a-workspace)
3. **Ask a question** — Type a question in the workspace chat bar

For the best results, enrich your models with `#(doc)` descriptions and `#(index)` annotations — see [Metadata Tags](/how-to/modeling/metadata-tags).

## See Also

* [Quick Start](/quickstart-build) — End-to-end tutorial including workspace chat
* [Credible Context Engine](/concepts/context-engine) — How phrase-matching and entity retrieval work
* [MCP Reference](/reference/api/mcp) — Technical details on `get_context` and `execute_query`
* [Credible App Admin](/platform-admin/admin-portal) — Creating and managing workspaces
