> For the complete documentation index, see [llms.txt](https://docs.spara.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.spara.com/agents/channels/configuring/ask-spara-ai.md).

# Ask Spara AI

**Ask Spara AI** is an AI assistant built into the Spara platform that helps you understand and improve your channels. Ask it questions about your prompt, and it responds with analysis, specific line citations from your instructions, and edits you can apply with a single click, without leaving the editor.

<figure><img src="/files/ciXbif1uQ815PkVAQF9N" alt=""><figcaption><p>Ask Spara AI in the editor's right-hand panel.</p></figcaption></figure>

## Where to find it

Ask Spara AI is available in two places:

**In the Chat editor**: accessible in the right-hand panel. Toggle between the **Testing** interface and **Ask Spara AI** using the tab at the top of the panel. This lets you read and edit the prompt on the left while getting AI feedback on the right.

**In the Phone test panel**: the [Phone](/agents/channels/phone.md) channel has a dedicated **Ask Spara AI** tab in its test panel, alongside the other testing options. This gives you prompt analysis in the same place you test calls.

## What you can ask

Ask Spara AI can answer questions across several categories:

**Diagnosing behavior**

* "Why did it respond this way?" references the most recent test conversation
* "Find examples in the test conversation where it didn't follow the prompt"
* "Review the tools configured on this channel"

**Auditing the prompt**

* "Check for conflicting instructions"
* "Are there any gaps in how I've handled objections?"
* "Summarize the instructions"

**Improving the prompt**

* "How do I make it respond with shorter, friendlier sentences?"
* "Rewrite the qualification section to be more direct"
* "Add instructions for handling pricing questions"

## How edits work

When Ask Spara AI suggests a change to your prompt, it shows an **edit card**: a diff-style preview of the proposed change alongside the relevant section of your prompt. Click **Apply** to accept the edit directly into your draft. You can accept or dismiss each suggestion independently.

Applying an edit does not publish the channel; it saves to your draft. Review the result in the testing interface, then publish when you're satisfied.

## FAQ

### Does Ask Spara AI have access to my lead conversations?

In the editor, Ask Spara AI has access to the active test conversation on the right-hand panel. It uses that context when you ask questions like "why did it say that?" For real lead conversations, Ask Spara AI does not have access; it works from the prompt and the test session only.

### Can I ask it to rewrite the whole prompt?

Yes, but starting with targeted questions generally produces better results. Ask it to audit a specific section, apply the edit, test the result, and repeat, rather than requesting a full rewrite in one go.

### Are Ask Spara AI conversations saved?

The Ask Spara AI conversation is session-based and resets when you navigate away from the editor. Your applied edits are saved to the draft, but the chat history is not.


---

# 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://docs.spara.com/agents/channels/configuring/ask-spara-ai.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.