# Build

## Persona & Settings

We recommend starting with the default persona, then tuning the settings in order to get the behavior you want. These instructions can do everything from requesting a different tone of voice to allowing the agent to ask follow up questions.

<figure><img src="/files/9lxHLmLNxzFwLVUVyMxy" alt=""><figcaption></figcaption></figure>

## Training

Agents can be trained with [Resources](/pylon-docs/ai-agents/build/resources.md) or [Runbooks](/pylon-docs/ai-agents/build/runbooks.md).

Resources are forms of knowledge including your Pylon knowledge base, external web pages, or files.

Runbooks are instruction manuals for how the AI agent can handle various specific scenarios. Runbooks also allow the AI agent to take action.

## Agent Workflows

### Assignment

To assign issues to an agent, visit the Assignment tab of its configuration. There are two toggles here:

* Agent Enabled should always be toggled if you want to the AI agent to be able to perform actions when assigned an issue.

Once you've set up these toggles, you can set up rules via [Triggers](/pylon-docs/platform/triggers.md) just like you would for any other human teammate or manually assign relevant issues on a one-off basis.

Common assignment triggers include:

* Assign AI agent to low tier customer issues
* Assign AI agent to off-hours issues

### Greeting

Configure what your AI Agent should do when it first responds to a customer request.

### Resolution

Configure what your AI Agent should do when the user indicates the agent has answered their request.

### Escalation

Configure what your AI Agent should do when it does not have an answer to a customer request.


---

# 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.usepylon.com/pylon-docs/ai-agents/build.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.