# Saved Prompts

Saved Prompts are a shared library of your team's proven, go-to queries for a specific AI Analyst. Instead of retyping the same question every time — or trying to remember exactly how you phrased a question that produced a great result — [Admins](/settings/members.md) can save those prompts once and make them available to everyone in the organisation.

{% hint style="info" %}
Saved Prompts are org-wide and scoped to a specific AI Analyst. Everyone in your workspace can access and run them. Only Admins can create, edit, or delete them — see [Managing Saved Prompts](/ai-analysts/creating-an-agent/saved-prompts.md) for the Studio side.
{% endhint %}

## The Explorer homepage

When you open an AI Analyst in Explorer, the homepage shows two tabs:

* **Get started** — AI-generated suggestions to help you discover what the agent can do (great for first-time users)
* **Saved prompts** — Your team's curated library of proven queries for this agent

Click the **Saved prompts** tab to browse the list. Each entry shows the prompt title, the full prompt text, who saved it, and when. If the prompt was saved from a completed conversation, a **View original task** link lets you preview the result before re-running it yourself.

Click any saved prompt to immediately start a new conversation with that prompt pre-filled.

{% hint style="info" %}
If the agent has no saved prompts yet, the tab will show: *"No saved prompts yet. Admins can save prompts from any completed conversation."*
{% endhint %}

## The / trigger — access prompts mid-conversation

You don't have to go back to the homepage to use a saved prompt. While in any conversation, type `/` in the chat input to open the saved prompts dropdown.

* Start typing after `/` to filter by title or prompt text (e.g. `/revenue` narrows to revenue-related prompts)
* Use the arrow keys to navigate, then press **Enter** to select a prompt
* The prompt text is inserted into the input — you can edit it before sending
* Press **Escape** to close the dropdown without selecting

## Variable placeholders

Saved prompts can contain placeholders like `[region]` or `[timeframe]`. When a prompt with placeholders is selected, the placeholder text is highlighted so you can quickly fill it in before sending. Press **Tab** to jump between placeholders.

For example, a prompt like *"Show me revenue by \[region] for \[timeframe]"* lets anyone on the team run a region- and period-specific analysis without typing the full question from scratch.

## Saved Prompts vs. Suggestions

Both appear on the Explorer homepage, but serve different purposes:

|                       | Suggestions                                | Saved Prompts                                     |
| --------------------- | ------------------------------------------ | ------------------------------------------------- |
| **Created by**        | AI (auto-generated from semantic layer)    | Admins (human-curated)                            |
| **Purpose**           | Discovery — *"here's what you could ask"*  | Retrieval — *"here's what we know works"*         |
| **Prompt length**     | Short (up to 8 words)                      | Any length; can include `[variable]` placeholders |
| **Stability**         | Regenerate when the semantic layer changes | Stable until an Admin edits or deletes them       |
| **Linked to results** | No                                         | Optionally linked to the original task            |

Use Suggestions to help new users explore; use Saved Prompts to codify the queries your team runs repeatedly.


---

# Agent Instructions: 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:

```
GET https://docs.wobby.ai/explorer/working-with-agents/saved-prompts.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.