> ## Documentation Index
> Fetch the complete documentation index at: https://ionet-cca8037f-confidential-compute.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Exploring AI Agents

> Learn about AI Agents on io.net and how to use, configure, and deploy them to build autonomous AI systems.

Agents on [io.net](http://io.net) are packaged, executable components that encapsulate task logic, tool interactions, and external integrations. This section documents the agents provided by [io.net](http://io.net), including core agents and agents that integrate with third-party services such as *GitHub*, *Linear*, and other external tools. It describes how to deploy agents programmatically or via `curl`, configure runtime parameters, and manage secrets required for authenticated access to external APIs. These capabilities support repeatable, secure operation of AI agents within automated workflows.

## Getting Started

<Steps>
  <Step title="Navigate to the Agents tab">
    Open [io.net Intelligence](https://ai.io.net/) and navigate to the Agents tab in the navigation bar.

    <Frame>
      <img src="https://mintcdn.com/ionet-cca8037f-confidential-compute/MNuf6kXmoHAgX-oq/images/docs/io-intelligence/Agents/IOIntel_Agents_NavBar.png?fit=max&auto=format&n=MNuf6kXmoHAgX-oq&q=85&s=2eb202a71b1e33f66e89d4ba57eb2e2f" alt="Agents Tab in the Navigation Bar" width="1154" height="212" data-path="images/docs/io-intelligence/Agents/IOIntel_Agents_NavBar.png" />
    </Frame>
  </Step>

  <Step title="Select an Agent">
    Browse the list of available agents, or use the search bar to quickly locate a specific agent, then click on your chosen agent to view its details.

    <Frame>
      <img src="https://mintcdn.com/ionet-cca8037f-confidential-compute/MNuf6kXmoHAgX-oq/images/docs/io-intelligence/Agents/IOIntel_Agents_MainView.png?fit=max&auto=format&n=MNuf6kXmoHAgX-oq&q=85&s=6b1d942450088bd925c490f11a3441c6" alt="Agents Selection and Search Bar" width="2488" height="1620" data-path="images/docs/io-intelligence/Agents/IOIntel_Agents_MainView.png" />
    </Frame>
  </Step>
</Steps>

## **Deploying and Configuring Agents**

When an agent is selected, the *Agent Details* view provides everything needed to evaluate, install, and configure your chosen agent.

### Agent Details and Installation

The *Agent Details* view displays information about your selected agent, along with installation options. Agents can be installed using either Python (via `pip`) or `cURL`, depending on your preferred workflow.

If you do not already have an existing API key, the *Agent Details* view provides an option to generate and inject a new API key directly into the provided `pip` or `curl` installation commands, allowing installation to proceed without leaving the page.

<Frame>
  <img src="https://mintcdn.com/ionet-cca8037f-confidential-compute/MNuf6kXmoHAgX-oq/images/docs/io-intelligence/Agents/IOIntel_Agents_InjectAPIKey.png?fit=max&auto=format&n=MNuf6kXmoHAgX-oq&q=85&s=b1893097e25dffc29da1da9da5d22d5d" alt="IO Intel Agents Inject API Key" width="2338" height="1064" data-path="images/docs/io-intelligence/Agents/IOIntel_Agents_InjectAPIKey.png" />
</Frame>

### Trying an Agent before Installation

Before installing, an agent can be tested directly from the interface. Click the **Try Agent** button and it opens the *Playground* where you can be provide input for the agent.

After clicking the **Run Agent** button, the **Result** tab displays the agent’s output, making it possible to quickly validate behavior and responses before deployment.

<Frame>
  <img src="https://mintcdn.com/ionet-cca8037f-confidential-compute/MNuf6kXmoHAgX-oq/images/docs/io-intelligence/Agents/IOIntel_Agents_Playground.png?fit=max&auto=format&n=MNuf6kXmoHAgX-oq&q=85&s=5f6624841a36579fc66e9917d31ad289" alt="IO Intel Agents Playground" width="2464" height="1366" data-path="images/docs/io-intelligence/Agents/IOIntel_Agents_Playground.png" />
</Frame>

### Configuring Secrets

Agents that integrate with third-party applications require secrets for authenticated access, for example, GitHub, Linear, Jira and many more.

To configure secrets, navigate to the **Secret Management** tab within the *Agent Details* view.

<Frame>
  <img src="https://mintcdn.com/ionet-cca8037f-confidential-compute/MNuf6kXmoHAgX-oq/images/docs/io-intelligence/Agents/IOIntel_Agents_SecretManagement.png?fit=max&auto=format&n=MNuf6kXmoHAgX-oq&q=85&s=2a6476771cf4164038cffcd4e674f381" alt="IO Intel Agents Secret Management" width="2380" height="708" data-path="images/docs/io-intelligence/Agents/IOIntel_Agents_SecretManagement.png" />
</Frame>

On this tab:

* Clicking the **Manage Secrets** button redirects you to the *API Keys and Secrets* page.
* On the *API Keys and Secrets* page, a new secret can be added here by clicking **Add Secret** and providing the following details:
  * *Secret Name*
  * *Secret Value*
  * *Associated Agent*

Alternatively, secrets can be created or selected directly within the **Secret Management** section of the *Agent Details* view:

* An existing secret can be selected from the list.
* A new secret can be created by entering a secret name and value.
* Secrets created here are automatically saved and will also appear in the *API Keys and Secrets* page.

Once the required secrets are configured, the agent is fully set up and ready to be interacted with.