# File Reader

File Connectors allow you to structure and manage document sets that will serve as a knowledge base (RAG source) so that the AI Agent can respond contextually and accurately to user requests.

<figure><img src="/files/WVGjjuoqIbCkENq5IJf3" alt=""><figcaption></figcaption></figure>

### 1. Workflow Overview

The process to grant an Agent access to documents follows three steps:

1. **Upload:** Add files in **Artifacts > Files**.
2. **Connector:** Create the Connector to group the relevant files.
3. **Relationship:** Associate this Connector with the specific AI Agent.

### 2. Step by Step: Creating the File Connector

The Connector is created in the Administration Portal.

#### 2.1 Accessing the Connectors Area

<figure><img src="/files/0f9tj6RDWGveeBgDqaQe" alt=""><figcaption></figcaption></figure>

1. In the Administration Portal (**Artifacts**), navigate to **Connectors**.
2. Click **"Add Connector"** or the create button (usually a "+" icon).
3. Select the connector type: **"File Connector"**.

#### 2.2 Configuring the Connector

Fill in the required fields to name and select the documents that will be part of this set.

<figure><img src="/files/bjOsJp8N2Rx4eUs7YCfR" alt=""><figcaption></figcaption></figure>

* **Connector Name:** Enter a clear and descriptive name that identifies the file set (Ex: "HR Internal Policies", "Product X Documentation").
* **Project:** Select a project for your connector.

#### 2.3 Selecting Files

<figure><img src="/files/xKBnBwzcNyYWHf80HUHZ" alt=""><figcaption></figcaption></figure>

In the file selection section, you will find a list of all folders and documents previously uploaded in **Artifacts > Files**.

1. Locate and select the desired files that should compose this Connector. You can filter by folder or use the search bar to find specific documents.

<figure><img src="/files/U5o5MgbxbDD5uX1Wpr3r" alt=""><figcaption></figcaption></figure>

2. All selected files will be grouped together.

#### 2.4 Finalization

1. Review the settings and selected files.

<figure><img src="/files/mGVtg58J0uLO68XS0KB1" alt=""><figcaption></figcaption></figure>

2. Click **"Save"** to complete the creation of the File Connector.

{% hint style="info" %}
**Note:**&#x20;

Creating the Connector does not automatically grant permission to the Agent. The Agent must still be explicitly associated with this Connector.
{% endhint %}

### 3. Final Step: Linking the Connector to the Agent

This is the crucial governance step, where you define which Agents can access the new document set.

1. Go to **Artifacts > Agents** in the Administration Portal.

<figure><img src="/files/zvaZGsLmu9cuzNyIgyUY" alt=""><figcaption></figcaption></figure>

2. Find or create the desired AI Agent and click the relationships icon.

<figure><img src="/files/3HIfmomcYe0o1XdHT8tm" alt=""><figcaption></figcaption></figure>

3. Select **Connectors**.
4. In the list of available Connectors, locate the File Connector you just created (Ex: "HR Internal Policies").

<figure><img src="/files/caTIA2XyQkkuKA3unXQC" alt=""><figcaption></figcaption></figure>

5. Add or select the Connector to associate it with the Agent.
6. Click **"Save"** to apply the changes to the Agent configuration.

{% hint style="info" %}
It is important to adjust the AI Agent’s instructions so that it is aware that files are available for consultation and uses them as the basis for its responses.
{% endhint %}

#### ✅ Expected Result

Only after this final step, when a user interacts with the Agent, will the AI model have permission to retrieve information and generate responses using exclusively the files contained in this Connector and any other related Connectors.


---

# 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.poweromni.ai/en/administration-and-configuration/artifacts/connectors/file-reader.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.