> For the complete documentation index, see [llms.txt](https://docs.riff.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.riff.ai/archive/quick-start-tutorials-legacy-staging/quickstart-agent.md).

# Quickstart: Agent

You'll get an agent that searches the web, scores relevance using your criteria, and sends you a personalized digest. No feeds. No noise.

**Perfect for:** Knowledge workers drowning in newsletters, operators who need signal over noise, anyone building domain expertise.

### Features

* **Personalized search**: Agent decides which sources to search based on your topics
* **Smart filtering**: Scores articles against your job context and values
* **Autonomous scheduling**: Runs every Monday at 07:00 without you
* **Slack delivery**: Digest arrives in your [Slack ](/archive/integrations-legacy-saas-connectors/native-integrations/slack.md)channel with article summaries

### Create your news digest agent

{% stepper %}
{% step %}
**Step 1 - Create an account**

Go to **riff.ai** and sign up.
{% endstep %}

{% step %}
**Step 2 - Describe your agent**

Paste this into the prompt box:

> Build a personalised newsletter agent that uses OpenAI and Tavily MCP to search the internet for latest news and insights from my preferences. Compiling a daily newsletter that will be summarised and sent to Slack.
> {% endstep %}

{% step %}
**Step 3 - Confirm the plan → Build**

Riff will generate a Riff Card — the build plan. Review it. Adjust if needed. Click **Confirm plan → Build project**.
{% endstep %}

{% step %}
**Step 4 - Complete the tasks**

Your workspace is live.

You'll see a preview tab, this is the interface the agent has designed for your app.

{% hint style="info" %}
Note: This is a placeholder interface and will not function at this stage.
{% endhint %}

The agent creates a task list and runs until it needs input. Click on the [Tasks ](/archive/features-legacy-databutton-web-app-builder/workspace-features/tasks.md)tab to see the plan the agent has created to complete your application.

{% hint style="info" %}
Note: your tasks may look different.
{% endhint %}

* **Task 1** is already complete as the agent has created a user interface page.
* **Task 2** will typically be to set up your integrations for OpenAI, Tavily and Slack.
* **Task 3** create a backend for your app, this will be the process that will run. In this case, it will be searching and summarizing relevant content.
* **Task 4** is to make your personalise your agent search. In this case, the agent will:
  * Create an interface for you to enter your content
  * Set up the [database](/archive/features-legacy-databutton-web-app-builder/native-database.md) to store and retrieve your content preferences.
* **Tasks 5 & 6** create the [schedule ](/archive/features-legacy-databutton-web-app-builder/workspace-features/schedules.md)to trigger the process and deliver the outputs to [Slack](/archive/integrations-legacy-saas-connectors/native-integrations/slack.md).

The agent will ask for your **OpenAI API key and your Tavily MCP key**.
{% endstep %}

{% step %}
**Step 5 - Test your agent**

Click **"Run now"** to trigger the agent manually.

Check your Slack channel. You should receive a digest with 5 articles, scored and summarized based on your criteria.

**Not relevant enough?** Tell the agent in plain English:

* "Make the 'practical insights' filter stricter"
* "Ignore anything from corporate blogs"
* "Add 'AI regulation' as a topic"

The agent updates its logic. Test again.
{% endstep %}

{% step %}
**Step 6 - Deploy**

When it looks right: **Deploy** → choose a short URL → **Deploy app**.

Your agent now runs autonomously every Monday at 07:00. You can share the URL with teammates to build their own personalized digests.
{% endstep %}

{% step %}
**Well done!**

You've built an autonomous agent that researches, evaluates, and reports — without you lifting a finger!
{% endstep %}
{% endstepper %}

<details>

<summary>Get your Open AI API key</summary>

1. **Create an OpenAI Account**:

* Visit the OpenAI platform website and click on **Sign Up** if you don't have an account. You can also log in using your Google or Microsoft account if you prefer.

2. **Log In**:

* If you already have an account, click on **Log In** and enter your credentials.

3. **Access the API Keys Section**:

* Once logged in, click on your profile icon at the top-right corner of the page and select **View API Keys** from the dropdown menu.

4. **Create a New API Key**:

* On the API keys page, click on the **Create New Secret Key** button. This will generate a new API key for you.

</details>

<details>

<summary>Get your Tavily API Key</summary>

To **obtain your Tavily API key**, follow these steps:

* Go to the **Tavily website** (<https://tavily.com/>) and log in to your account. If you don’t have an account, sign up first.
* Navigate to the **API section** at <https://app.tavily.com/home>. Your API Key will be displayed in the “API Keys” section.
* If you need to create an additional API key, click on the "+" button next to the API keys section in your dashboard.
* You receive **free API Credits** every month, which allows for personal use and experimentation.

</details>

**Tip:**

* When prompted for your OpenAI key, get one at [platform.openai.com/api-keys](https://platform.openai.com/api-keys). Paste it in. The agent continues building.
* When prompted for your Tavily API key, get one at [app.tavily.com/home](https://app.tavily.com/home) Paste it in. The agent continues building.


---

# 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.riff.ai/archive/quick-start-tutorials-legacy-staging/quickstart-agent.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.