# Run a specialist agent

BriefTech includes a specialist AI agent designed for non-disclosure agreements. It is pre-configured with field definitions, extraction prompts, and few-shot examples tuned for that document class.

***

## Available specialist agent

| Agent         | Best for                                              | Key fields extracted                                                                 |
| ------------- | ----------------------------------------------------- | ------------------------------------------------------------------------------------ |
| **NDA agent** | Non-disclosure agreements, confidentiality agreements | Parties, effective date, term, definition of confidential information, governing law |

***

## When to run the NDA agent

Run the NDA agent after the Organise agent has run and produced initial summaries. The NDA agent uses those summaries as additional context to improve extraction accuracy.

Only run the NDA agent when your document set contains non-disclosure or confidentiality agreements.

***

## How to run the NDA agent

1. In the Review or AI panel, click **Run AI** → **NDA**.
2. *(Optional)* Apply a collection or folder filter first so the agent only processes the relevant documents.
3. Review the estimated credit cost in the confirmation dialog.
4. Click **Run**. Progress is shown in the **Background jobs** panel.

> The NDA agent only processes documents whose category matches its document type. If a document's category has not been set (or was set incorrectly), the agent may skip it — check and correct categories in the Review tab first.

***

## Reviewing specialist agent output

Once the job completes, extracted fields appear in the **Facts tab** with the agent name as the source. Review, confirm, or edit them as described in [Review and confirm extracted facts](/tutorials/tutorials/set-3-ai-analysis/02-facts-tab.md).

***

## What's next?

Continue to [Find and remove duplicates](/tutorials/tutorials/set-3-ai-analysis/04-find-remove-duplicates.md) to clean up your document set before building the LOD.

***

## Related resources

* [Run the Organise agent](/tutorials/tutorials/set-3-ai-analysis/01-run-organise-agent.md) — tutorial — prerequisite: run Organise first for best results
* [Review and confirm extracted facts](/tutorials/tutorials/set-3-ai-analysis/02-facts-tab.md) — how-to — review the specialist agent's output here
* [Credits and billing overview](/tutorials/tutorials/set-8-billing/01-credits-billing.md) — concept — credit cost per page processed


---

# 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://help.brieftech.ai/tutorials/tutorials/set-3-ai-analysis/03-specialist-agents.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.
