# Creating an Agent

Custom agents let you define exactly how the AI behaves — its role, expertise, tone, and the knowledge it draws from. Once created, an agent is available across all your chats and can be shared within your workspace.

***

#### Step 1 — Open the Agent Creation Panel

1. Go to the **chat input area**
2. Click **Create Agent**

This opens the agent creation panel.

***

#### Step 2 — Select a Base Model

Choose the AI model your agent will run on. Consider:

* **Reasoning capability** — for analytical or research-heavy agents
* **Creativity** — for content or creative roles
* **Context length** — for agents handling long documents
* **Credit consumption** — for cost efficiency at scale

> **Important:** If you plan to attach files or knowledge bases to your agent, select a **tool-calling enabled model**. Resource attachment requires tool support.

***

#### Step 3 — Name Your Agent

Give your agent a clear, descriptive name that reflects its role.

**Examples:**

* `SaaS Marketing Strategist`
* `Retail Analytics Advisor`
* `Legal Compliance Reviewer`

Optionally, add a **tagline** — a one-line description of what the agent does. Example: *"Expert advisor for B2B SaaS growth and customer acquisition strategy."*

***

#### Step 4 — Add Resources (Optional)

Resources give your agent persistent access to structured knowledge — files and documents it references automatically in every chat.

**To enable:** Turn on the **Add Resources** switch.

You can attach resources in two ways:

**From existing Knowledge Bases** — Select any Knowledge Base you have already created and attach it to the agent. See Knowledge Bases → for how to create and manage them.

**Upload new files** — Upload documents directly to the agent.

Supported formats: PDF, CSV, DOC, DOCX, Excel, TXT

| File Type | Size Limit                  |
| --------- | --------------------------- |
| Images    | Max 20 MB                   |
| Documents | Up to 1,000 pages or 200 MB |

***

#### Step 5 — Write Agent Instructions

This is the most important step. Instructions define your agent's identity, role, expertise, and behavior — they become the system prompt the model follows in every response.

Structure your instructions across four areas:

1. **Background** — Define the professional identity.

```
You are a senior marketing strategist with 10+ years of
experience in SaaS and consumer tech.
```

2. **Role** — Specify what the agent should do.

```
Provide structured insights, campaign strategy,
and performance analysis.
```

3. **Expertise** — Define domain specialization.

```
Expert in digital acquisition, customer psychology,
and data-driven growth.
```

4. **Behavioral Instructions** — Define tone and output style.

```
Be concise and structured.
Provide step-by-step explanations.
Include actionable recommendations.
Avoid unnecessary filler.
```

{% hint style="info" %}
The more specific your instructions, the more consistent and useful your agent's outputs will be. Vague instructions produce generic behavior — treat this like writing a detailed job description.
{% endhint %}

***

#### Step 6 — Create the Agent

Once you have completed all steps, click **Create Agent**. Your agent is now available in the Agent Selector and ready to use in any chat.


---

# 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.qolaba.ai/chatbot/agents/creating-an-agent.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.