> 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/playbook.md).

# Playbook

## What is Playbook?

Playbook is a library of natural-language protocols that teach the AI assistant how your firm - and you personally - want each task handled. Every protocol you save is applied automatically whenever the assistant runs that type of task. Unlike instructions you type into a Chat session, protocols persist across sessions and clients: you write a rule once, and the assistant follows it every time without you having to repeat it.

## How protocols work

Each protocol has two parts:

* **Description (the trigger)** - tells the system what situation this protocol applies to. The assistant reads it to decide whether the protocol is relevant for the current run.
* **Body (the instruction)** - the actual guidance the assistant follows when the protocol is active, written in plain language with as much or as little detail as the rule requires.

When a task starts, the assistant automatically loads every active protocol for that task type and reads them as part of its working context. No manual step is needed at run time. The protocols are injected into the assistant's prompt as a structured block, so the guidance is available from the first action the assistant takes.

## The five task types

Protocols are organized by the type of work they govern. A protocol created for one task type will not run during any other.

* **AI Tax Prep** -- Protocols that shape how the AI Tax Prep assistant fills returns, handles specific schedules, and resolves data-entry decisions.
* **AI Review** -- Protocols that influence how the AI Review assistant applies check rules, flags findings, and writes leadsheet observations.
* **AI Tax Planning** -- Protocols that guide AI Tax Planning when identifying strategies and preparing planning materials.
* **Binder** -- Protocols that affect how the assistant handles document processing, categorization, and binder organization.
* **Chat** -- Protocols that carry persistent preferences into general Chat sessions.

## Your protocols vs. your firm's protocols

Every Playbook tab shows two sections, "Yours" and "Firm's":

* **Yours** are personal protocols. Only you can see them, and they run only during your own task sessions. Use them to encode your individual working style, preferences for certain client types, or rules you want to test before sharing more broadly.
* **Firm's** are shared protocols. They run for every preparer at the firm during the relevant task type, and are either curated defaults shipped by Filed or personal protocols that an admin has promoted to firm scope.

See [Yours vs. Firm's Protocols](/reference/playbook/yours-vs-firms-protocols.md) for details on scope, promotion, and admin governance.

## Where Playbook lives in the app

Playbook appears as a dedicated entry in the left-side navigation. Clicking it lands you on the AI Tax Prep tab by default. The left sidebar inside Playbook lists all five task types in order: AI Tax Prep, AI Review, AI Tax Planning, Binder, Chat. Clicking a task type shows its protocol list.

Each task-type page shows two sections stacked vertically: Yours at the top, Firm's below. Each row shows the protocol name, its one-line description (the trigger), the author, and a status badge. Clicking a protocol name opens its detail view, which shows the full body text and an activity timeline recording every lifecycle event.

## Why protocols beat one-off instructions

There are three ways to give the assistant guidance, and they behave very differently:

* **Typing an instruction in Chat** works for that session only. Once the conversation ends, the instruction is gone, and the assistant starts fresh on the next client.
* **A personal protocol in Yours** persists permanently. Every time you run that task type, the assistant applies it automatically - whether you are on your first client of the day or your fiftieth.
* **A firm protocol in Firm's** does the same, but for every preparer at the firm. Consistency that would otherwise depend on everyone remembering to mention the same thing in Chat is enforced automatically at run time.

For a practice with multiple preparers and many clients, the compounding effect is significant: the assistant's behavior becomes predictable and consistent across the whole team without any manual coordination.

## In this section

* [Creating Protocols](/reference/playbook/creating-protocols.md) -- How to author a new personal protocol through Chat and confirm it is active in the Playbook list.
* [Yours vs. Firm's Protocols](/reference/playbook/yours-vs-firms-protocols.md) -- Scope, ownership, the promotion workflow from personal to firm, and admin governance.


---

# 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/playbook.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.
