> 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/context.md).

# Context

Context is what you tell Coupler AI about your data and your business so it interprets both correctly. It comes in three layers: data context, business context, and skills.

An AI tool connected to raw data can read column names and run queries. What it can't do is know that `revenue` excludes refunds, that your fiscal quarter starts in February, or that last March's spike was a pricing change and not a trend.

Context is the knowledge you add so it doesn't have to guess. You add it once, it's stored with your data and your workspace, and every conversation after that starts informed, whether it happens in the AI Agent or in an AI tool you've connected.

## The three layers

| Layer                                                   | What it holds                                                                                                                                                                    |
| ------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [**Data context**](/ai/context/data-context.md)         | What the data means. What each data set represents, what every column holds, its type and units, and the business names for your metrics and dimensions.                         |
| [**Business context**](/ai/context/business-context.md) | How to interpret it. The logic behind the numbers, what one row represents, the gaps to account for, your calendar of launches and outages, and the targets you measure against. |
| [**Skills**](/ai/context/skills.md)                     | How the work gets done. Step-by-step procedures for analyses that repeat, including the output format and the checks to run before an answer is delivered.                       |

They stack. Data context lets the AI pick the right columns, business context lets it read the result correctly, and a skill turns that into the same analysis your team would run by hand.

## Where each layer lives

* **Data context** is stored on the data set, as column descriptions and schema details. Coupler.io detects it and provides it by default, so every data set has a semantic layer from the start, and you edit it to match the logic your data actually holds. The AI reads it before it queries anything.
* **Business context** is stored on the data set too, as a description you write or generate in the **Context** tab of the data set preview.
* **Skills** are stored in your workspace, so they apply across data sets rather than to one.

Because all three live with your data and your workspace instead of in a chat window, an update is picked up everywhere, including in a different AI tool and in conversations started by other people on your team.

## Where to start

{% stepper %}
{% step %}
**Refine the data context on the data set you query most.** Coupler.io already detected one, so go through it and correct what doesn't match your logic. This is the fastest improvement to answer quality.
{% endstep %}

{% step %}
**Add business context to the same data set.** Cover what one row represents, how the key metrics are defined, and any known gaps in the data.
{% endstep %}

{% step %}
**Add a skill for the analysis you run most often.** Once the same question comes up weekly, write down the procedure instead of re-explaining it.
{% endstep %}
{% endstepper %}


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.coupler.io/ai/context.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.