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

# Playbooks

A playbook is the answer to "I keep asking the same kind of question — can SMAQ just remember how I want it answered?"

A playbook bundles:

* A clear **name** and **description**.
* A curated set of **datablocks** the AI should consult.
* Free-text **guidelines** that shape tone, length, focus.

Once saved, the playbook appears at the top of chat suggestions whenever a related question is asked. One click and SMAQ runs your standard analysis your way.

> **\[Screenshot needed]** Playbook detail page showing the name, description, three attached datablocks, and the guidelines field.

### When playbooks pay off

* You ask the same question every week ("Weekly performance review").
* Your client expects a specific format ("Q2 spend efficiency report").
* You want the AI to write in a specific voice ("Casual tone, 200 words max, end with a recommendation").
* Multiple team members need to ask the same question consistently.

### Creating a playbook

See [Build your first playbook (tutorial)](/documentation/get-started/tutorials/build-your-first-playbook.md) for the step-by-step. The short version:

1. **Playbooks → Create new playbook**.
2. Name and describe.
3. Add datablocks (search → +).
4. Add guidelines.
5. Save.

> **\[Screen video needed]** Creating a playbook from scratch — name, description, add three datablocks, add guidelines, save, then execute from chat. \~2 minutes.

### Where the guidelines go

The guidelines field is free text. Examples that work well:

* **Tone**: "Write casually, like a Slack message between teammates."
* **Length**: "Keep responses under 200 words."
* **Format**: "Always end with three numbered next-step recommendations."
* **Language**: "Reply in 繁體中文."
* **Focus**: "Lead with what changed week-over-week, ignore overall averages."
* **Audience**: "Audience is a CMO, not an operator — skip platform-specific tactics."
* **Constraints**: "Never recommend pausing the Brand campaign."

The AI honors guidelines on every playbook run.

### Sharing playbooks

#### Org-wide sharing

Toggle **Org-wide sharing** on the playbook detail page. The playbook becomes available in **Organization Gallery** for every project in your org.

> **\[Screenshot needed]** Playbook with Org-wide sharing toggle on, plus the Organization Gallery showing it listed.

#### Copy a shared playbook

From any project:

1. **Playbooks → Organization Gallery** tab.
2. Find the shared playbook.
3. Click **Copy to project**.

A copy lands in **My Playbooks** in that project. You can edit it without affecting the shared original.

### Executing a playbook

#### From chat suggestions

When you open chat in a project that has playbooks, the chat suggestions strip surfaces them. Click to execute.

#### By asking related questions

If you ask a question that matches a playbook's description, SMAQ surfaces the playbook at the top of the response with a "Run this playbook?" prompt. Click yes.

> **\[Screenshot needed]** Chat with the AI suggesting a playbook at the top of its response — "I think the Weekly Performance Review playbook fits this — run it?"

### Tips

* **Start with one playbook.** Refine it over a week of use. Then add more.
* **Narrow > broad.** "Weekly Meta review" produces sharper output than "Full marketing review."
* **Embed your house style.** Reusing the same opener and closer across reports? Put it in the guidelines.
* **Keep datablocks tight.** 3–5 datablocks usually outperforms 10. The AI handles narrower contexts better.

### Tier availability

* **All tiers** — Playbooks are included starting at Starter.

### Related

* [Build your first playbook (tutorial)](/documentation/get-started/tutorials/build-your-first-playbook.md).
* [What SMAQ AI can do](/documentation/features/chat/capabilities.md) — see the Execute Playbook Agent in context.
* [Knowledge Base](/documentation/features/knowledge-base.md) — playbooks reference KB context automatically.