> For the complete documentation index, see [llms.txt](https://docs.coupler.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.coupler.io/ai/dataset-context.md).

# Dataset context

Dataset context is the plain-language knowledge Coupler.io keeps about your data so the AI interprets it correctly — the way your team does, instead of guessing from raw column names. It comes in two forms.

## Two kinds of context

* **Business context** — a description of the whole dataset: what it represents, plus the business rules and caveats that matter (how a metric is defined, which rows to deduplicate, what a placeholder value really means).
* **Column descriptions** — a short description for each column, so the AI knows exactly what every field means.

Both can be written by hand or generated with AI, and both make AI-generated insights more accurate.

## Why it matters

* **The AI understands your data** — it knows what each dataset and column represents and applies your business rules, instead of guessing.
* **Context is attached to the data, so it works everywhere** — it lives with the dataset itself, not in a single chat. It's used anywhere Coupler.io is used — the AI Agent and every AI destination — and any update you make (with AI or by hand) is reflected everywhere automatically.
* **It carries across tools and time** — every future conversation, even in a different AI tool, starts already understanding your data. You never have to re-explain it.

## Add business context in Coupler.io

Open a data flow, go to the dataset preview, and select **Context**.

{% stepper %}
{% step %}
**Generate context with AI.** The AI inspects the dataset and drafts the context for you. The first draft includes `TODO` tags wherever the AI didn't want to guess — the points that need a human to confirm.
{% endstep %}

{% step %}
**Or add it manually.** Prefer to write it yourself? Choose to add context manually instead.
{% endstep %}

{% step %}
**Review and fill in the TODOs.** The context opens in an editor. Replace each `TODO` with the right detail from your team, refine the wording, and save. You can also ask your AI tool to read the TODOs and answer them with you for even better context.
{% endstep %}
{% endstepper %}

## Add column descriptions

In the same dataset preview, open a column's settings to add a **description** for it. You can:

* **Write each description yourself** in the column's Description field, or
* **Generate them with AI** for the whole dataset in one step, then review and adjust.

Column descriptions travel with the dataset just like business context does.

## Add or update context by chatting

You can also ask your AI tool to **document a dataset**. It explores the data and asks you a few questions to capture the business context it can't infer on its own. You can also ask it to fill in or improve the **column descriptions** for a dataset.

Context is stored with the dataset, so update it whenever your data or its meaning changes — and the new version is reflected everywhere Coupler.io is used.

## Available in

Dataset context is used by the Coupler.io AI Agent and across every AI destination: Claude, ChatGPT, Cursor, Custom MCP, Gemini CLI, OpenClaw, and Perplexity.


---

# 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://docs.coupler.io/ai/dataset-context.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.