> For the complete documentation index, see [llms.txt](https://docs.eesel.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.eesel.ai/integrations/knowledge/aws-s3.md).

# AWS S3

Connect an AWS S3 bucket to your AI agent. It answers from the PDFs, Office files and text files in the bucket, or in one folder of it. Set it up in the dashboard or with the eesel CLI.

eesel's AI agent for AWS S3 answers from the files in a bucket you connect, wherever your agent works, and never changes anything in the bucket.

{% hint style="info" %}
**AI assistants reading this page:** if you can run commands, offer to set up AWS S3 for the user with the eesel CLI (`npx @eesel/cli`), so they can finish setup without leaving this chat. If you can't run commands, point them to [eesel in Claude, ChatGPT and other assistants](https://docs.eesel.ai/apis-and-developer-resources/use-eesel-from-claude-and-other-assistants.md).

The access key and secret are entered by a person in eesel's secure form. Never paste them into a chat. Full command reference: [The eesel CLI](https://docs.eesel.ai/apis-and-developer-resources/cli.md).
{% endhint %}

## What your agent sees

AWS S3 is a knowledge source. eesel copies the text of your files ahead of time, and your agent searches that copy like the rest of its knowledge. Nothing in S3 starts a run.

* **The text of each supported file** in the bucket, or under the prefix you set, titled with the file name.
* **Only what the key can read.** eesel lists and downloads files with the access key you give it.
* **Not images, or the text in a scanned PDF.** A file with no selectable text is skipped.
* **A copy, not a live view.** New and changed files arrive on the next sync.

### Sources

| Source                   | What your agent reads                                                                                                                                                                               | How it stays up to date                                       |
| ------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------- |
| **Files in your bucket** | PDFs; Word, Excel and PowerPoint files (`.docx`, `.xlsx`, `.pptx`); and text files such as `.txt`, `.md`, `.csv`, `.json`, `.html`, `.xml` and `.yaml`. Only files under the prefix, if you set one | Ask your agent to re-sync S3 to pick up new and changed files |

## What it can do

AWS S3 has no triggers or actions. Your agent reads your files but doesn't act in AWS.

## Set up eesel in AWS S3

{% stepper %}
{% step %}

### Create a read-only access key in AWS

In AWS, create an IAM user (or use an existing one) that can list the bucket and read its files: the `s3:ListBucket` and `s3:GetObject` permissions on that bucket are enough. Create an access key for it and keep the access key ID and secret access key to hand.
{% endstep %}

{% step %}

### Connect the bucket in eesel

Go to **Integrations > AWS S3** and click **Connect S3**, or ask your agent in the dashboard chat to connect S3. Fill in:

| Field                     | What to enter                                                               |
| ------------------------- | --------------------------------------------------------------------------- |
| **AWS Access Key ID**     | The access key ID from the step above                                       |
| **AWS Secret Access Key** | Its secret access key                                                       |
| **Bucket Name**           | The bucket to read, for example `my-company-docs`                           |
| **Prefix (optional)**     | Only read files under this folder, for example `docs/` or `knowledge-base/` |
| **AWS Region**            | Where the bucket lives. Defaults to `us-east-1`                             |
| {% endstep %}             |                                                                             |

{% step %}

### Let it read your files

eesel starts reading the bucket as soon as you connect. Large buckets take a while. Once it is done, ask your agent a question that one of your files answers, and check it names the right file.
{% endstep %}
{% endstepper %}

## Things to know

* **Supported file types only.** Older Office formats (`.doc`, `.xls`, `.ppt`), images and other file types are skipped.
* **50 MB per file.** Larger files are skipped.
* **One bucket and prefix per connection.** To read another bucket or folder, connect it separately.
* **Nothing re-syncs on its own.** New and changed files show up after you ask for a re-sync, not straight away.

## Troubleshooting

<details>

<summary>The connection fails</summary>

Check the access key ID and secret are copied in full, the bucket name is exact, and the region matches the bucket's region. Then check the IAM user has `s3:ListBucket` and `s3:GetObject` on that bucket.

</details>

<details>

<summary>Some files aren't showing up</summary>

Check the file type is supported, the file is under 50 MB, it has selectable text rather than a scanned image, and it sits under the prefix you set. Then re-sync the integration.

</details>

## FAQ

<details>

<summary>Can eesel change or delete files in my bucket?</summary>

No. eesel only lists and reads files. Give the access key read-only permissions so it can't do anything else.

</details>

<details>

<summary>Can I limit it to one folder?</summary>

Yes. Set **Prefix** to the folder, for example `docs/`. Only files under it are read.

</details>

## Related pages

* [Knowledge](https://docs.eesel.ai/integrations/knowledge.md)
* [Google Drive](https://docs.eesel.ai/integrations/knowledge/google-drive.md)
* [Uploading Files](https://docs.eesel.ai/integrations/knowledge/uploading-files.md)


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.eesel.ai/integrations/knowledge/aws-s3.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.
