> For the complete documentation index, see [llms.txt](https://help.filed.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.filed.com/reference/products/chat-and-analyst.md).

# Chat & Analyst

## What is conversational AI in Filed?

Filed embeds an AI assistant in multiple places across the product. All surfaces share the same underlying engine but differ in scope: per-client Chat is bounded to one client's documents and return data, while the firm-wide Analyst spans the entire workspace. Answers are grounded in real data from your binder documents, return outputs, and your firm's book of business, not generic knowledge.

## The three surfaces at a glance

| Surface               | Where to find it           | Scope                             | Best for                                                             |
| --------------------- | -------------------------- | --------------------------------- | -------------------------------------------------------------------- |
| Per-client Chat       | Client hub, Chat task card | One client's documents and return | Questions about a single client's tax position, documents, or return |
| Firm-wide Analyst     | Left navigation            | Entire workspace                  | Cross-client pipeline questions, aggregate queries, tax research     |
| Home "Ask a question" | Home page                  | Entire workspace                  | Quick entry point; opens a new Analyst thread automatically          |

## Per-client Chat

Per-client Chat appears as a task card on a client's hub once binder creation is complete. The assistant is scoped entirely to that client: it can read the binder documents, the prepared return, the tax position summary, review findings, and client notes. It cannot see any other client's data. See [Per-client Chat](/reference/products/chat-and-analyst/client-chat.md) for full details.

## Firm-wide Analyst

The Analyst is accessible from the left navigation and spans the entire workspace. It can answer questions that cross client boundaries, such as which clients are missing documents, who is likely to owe, and which returns are ready to file. It also supports general tax research using live web search. The Analyst maintains a persistent thread history. See [Analyst](/reference/products/chat-and-analyst/analyst.md) for full details.

## Home "Ask a question" box

The Home page includes a quick-entry box for asking a question. Submitting a message there navigates automatically to the Analyst and opens a new thread. Threads started from the Home box live in the Analyst's thread history; there is no separate Home conversation history.

## Choosing the right surface

**"I have a question about one specific client"** - Use per-client Chat. The assistant can cite binder documents and drill into the return.

**"I want to see which clients are missing docs, ready to file, or likely to owe"** - Use the Analyst. It queries your full book of business.

**"I want to look up a current tax rate, threshold, or IRS rule"** - Use the Analyst. It fetches live data from the web and cites its sources.

**"I'm on the Home page and have a quick question"** - Type it in the Home box. You will land in the Analyst with a new thread.

## In this section

* **Per-client Chat** - Ask questions grounded in one client's documents and return data.
* **Firm-wide Analyst** - Cross-client queries, pipeline overviews, and tax research.
* [**Chat features**](/reference/products/chat-and-analyst/chat-features.md) - Dictate, source citations, flags and notes that persist into the binder, and alternate output formats.
* [**Prompting best practices**](/reference/products/chat-and-analyst/prompting-best-practices.md) - Getting better answers, using dictation, requesting CSV or spreadsheet output, and choosing between Chat and Analyst.


---

# 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://help.filed.com/reference/products/chat-and-analyst.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.