> For the complete documentation index, see [llms.txt](https://v2.dataos.info/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://v2.dataos.info/consume/consume-with-ai/overview.md).

# Overview

Use Data Product MCP to connect MCP-compatible AI assistants and agentic frameworks to Data Products. Once connected, you can discover Data Products, ask business questions, inspect health and lineage, and identify the right owner for follow-up.

To start, pick the tool you already work in. From the Data Product page, select **Activate → MCP** to open the **Connect with MCP** page, where you choose a supported AI client or agentic framework to wire up to governed data — no separate data tooling to learn.

![Connect with MCP page showing supported AI clients and frameworks](/files/ZdcrNmRfj7vn4YoGMNnG)

## How Data Product MCP works

Data Product MCP exposes Data Products to AI tools through the [Model Context Protocol](https://modelcontextprotocol.io). The AI assistant does not get an unrestricted database connection. It calls Data Product MCP tools, and those tools route through Data Product APIs, authorization, semantic definitions, quality checks, lineage, and ownership metadata.

This gives data consumers a single entry point:

<table><thead><tr><th width="127.43988037109375">Need</th><th>What Data Product MCP helps with</th></tr></thead><tbody><tr><td>Find</td><td>Locate Data Products by topic, domain, owner, metric, or business question.</td></tr><tr><td>Trust</td><td>Review semantic fields, quality status, run history, lineage, table profile, limitations, and owner information.</td></tr><tr><td>Ask</td><td>Query governed metrics and dimensions through the Data Product semantic layer.</td></tr><tr><td>Act</td><td>Use failure details and owner context to follow up with the right team.</td></tr></tbody></table>

## Consumer journey with AI

The AI-assisted consumption experience follows four stages.

<table><thead><tr><th width="125.89874267578125">Stage</th><th>What the user asks</th><th>What Data Product MCP returns</th></tr></thead><tbody><tr><td>Find</td><td>"What Data Products do we have for supplier performance?"</td><td>Relevant Data Products or tables from the catalog, scoped to what the user is authorized to see.</td></tr><tr><td>Trust</td><td>"Is this data fresh enough to act on?"</td><td>Quality status, run history, semantic surface, lineage, table profile, limitations, and owner information.</td></tr><tr><td>Ask</td><td>"What was quarterly revenue by customer segment?"</td><td>A governed answer from the Data Product semantic layer, with the metric and source model cited.</td></tr><tr><td>Act</td><td>"Why did this query fail?"</td><td>A clear failure reason and, for query failures, the named Data Product owner to contact.</td></tr></tbody></table>

## Security and governance

Data Product MCP inherits Data Product governance. Every request carries the user's own credentials, and DataOS authorizes the request before returning data. The assistant sees only the Data Products, tables, columns, and values the user is allowed to access.

If a Data Product masks a column for the user, the AI assistant gets the masked value. If the user does not have access to a Data Product, Data Product MCP does not expose it through discovery, lineage, profile, or query responses.

## Trust model

The assistant does not invent the data. Data Product MCP translates the user's question into structured tool calls against Data Products and returns response envelopes that include the relevant metric, model, quality status, lineage, run, or owner fields.

For analytical answers, the response cites the metric used and the source Data Product or model. If the question cannot be answered from the available semantic surface, Data Product MCP returns a clear limitation instead of fabricating a measure.

## Next steps

* [Connect a client](/consume/consume-with-ai/connect-clients.md) to use Data Product MCP from Cursor, Claude, Copilot in VS Code, or Codex Desktop.
* [Connect an agentic framework](/consume/consume-with-ai/connect-agentic-frameworks.md) to use Data Product MCP from LangChain or Vercel AI SDK.
* [Discover with AI](/consume/discover/discover-with-ai.md) to find relevant Data Products and tables.
* [Understand with AI](/consume/understand/understand-and-trust-with-ai.md) to inspect schema, quality, lineage, profiles, and runs.

To know more about how Data Product MCP works and what trust boundaries apply, see the [AI Activation reference](/concepts/foundations/activation/ai-activation.md).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://v2.dataos.info/consume/consume-with-ai/overview.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.