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

# Using Paxton AI

Chat with Paxton inside a matter to draft, research, and analyze your files.

Chat with Paxton inside a matter to draft documents, research the law, and analyze your files. Every chat starts with the matter's files, jurisdiction, and earlier work already in context.

### Start a chat

1. Open the matter from your **Matters** page.
2. Check the jurisdiction in the **Matter Details** panel. Paxton researches in it by default.
3. Type your request in **What do you want to work on?** on the Matter Overview and send it. You can also click **New Chat** in the left sidebar.
4. Follow up in the same chat. Paxton keeps the context of the whole conversation.

<figure><img src="https://1404317777-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCY0Bz9NWp8rqtfuqg7ne%2Fuploads%2FTibqkXfuQAmV8cjbLqFr%2Fcomposer.png?alt=media&amp;token=0337386b-a971-4dc8-81a6-154eb941617a" alt=""><figcaption></figcaption></figure>

### Use the composer

| Control        | What it does                                                                                                                                                                                                                              |
| -------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **+**          | **Add Files**: upload a file or attach one from the matter. **Writing Sample**: give Paxton a document to follow in style, tone, and reasoning. See [Drafting in Paxton AI](https://help.paxton.ai/help/paxton-ai/drafting-in-paxton-ai). |
| **Research**   | Turn knowledge sources on or off: **Legal Research** (and its jurisdiction), **Web search**, and **Secondary sources**. See [Research in the AI Assistant](https://help.paxton.ai/help/paxton-ai/research-in-the-ai-assistant).           |
| **Skills**     | Run a saved set of instructions for a common task. See [Skills and Web Search](https://help.paxton.ai/help/paxton-ai/using-paxton-ai/skills-and-web-search).                                                                              |
| **Microphone** | Dictate your request.                                                                                                                                                                                                                     |

### What Paxton already knows

* **Matter files**: refer to a document by name, such as "the Vertex offer letter," instead of uploading it again.
* **Jurisdiction**: research follows the matter's jurisdiction. To look elsewhere for one question, say so in your prompt.
* **Research across every source**: Paxton's frontier research agents work across the full legal library, law review commentary, and the web. See [Research Coverage](https://help.paxton.ai/help/paxton-ai/using-paxton-ai/research-coverage).
* **Earlier work**: past chats and drafts stay with the matter.

To set how Paxton responds in every chat, add **Custom Instructions** in **Settings > Preferences**.

### What can I ask Paxton?

* **Draft**: "Draft Plaintiff's first set of interrogatories to Vertex Solutions on the restructuring."
* **Analyze files**: "Summarize the key takeaways from this 100-page deposition."
* **Research**: "What's the Texas standard for pretext under McDonnell Douglas?"
* **Find information**: "List every mention of the force majeure clause across the contracts in this matter."
* **Strategize**: "Outline three arguments for opposing summary judgment based on the facts in this matter."

For more, browse the [Prompt Library](https://help.paxton.ai/help/prompt-library/start-here).

### Where does my work go?

* **Drafts**: documents Paxton creates appear on the **Drafts** page and open in [Canvas](https://help.paxton.ai/help/paxton-ai/editing-drafts-with-canvas).
* **Matter Activity**: recent chats and drafts appear on the Matter Overview.
* **Chat History**: your chats are listed in the left sidebar.
* **Tables** and **Matter Records**: structured outputs built from the matter's files.

### Tips for better results

* **Keep the matter organized.** Paxton draws on the matter's files, so a clean, current file set gives sharper answers.
* **Name the document you mean.** "In the March 12 termination letter" beats "in the letter."
* **Be specific.** Add the document type, date range, or output format you want.
* **Iterate.** Follow up in the same chat instead of starting over.
* **One chat per task.** It keeps your history easy to search.

{% hint style="warning" %}
Review Paxton's output before you file or send anything, and check citations with the [AI Citator](https://help.paxton.ai/help/legal-research/citations-and-ai-citator).
{% endhint %}

### Related articles

* [Research in the AI Assistant](https://help.paxton.ai/help/paxton-ai/research-in-the-ai-assistant)
* [Drafting in Paxton AI](https://help.paxton.ai/help/paxton-ai/drafting-in-paxton-ai)
* [Uploading Files](https://help.paxton.ai/help/other-topics/uploading-files)


---

# 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://help.paxton.ai/help/paxton-ai/using-paxton-ai.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.
