> ## Documentation Index
> Fetch the complete documentation index at: https://docs.atako.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenAI

> Connect OpenAI to your Atako agents — 17 read and 17 write actions.

Let your agents use the OpenAI API — chat completions, responses, embeddings, images, moderation, files, fine-tuning, batches and vector stores.

## Connection

* **Authentication**: API key (API key (sk-...)).

<Note>
  Go to platform.openai.com → sign in → Settings (gear icon) → API keys → "Create new secret key". Pick your project, leave permissions on "All" (or restrict to the endpoints you need), then copy the key (starts with sk-). Make sure the project has billing enabled.

  See [OpenAI's documentation](https://platform.openai.com/api-keys).
</Note>

## Read actions (17)

| Action                      | Description                                                          |
| --------------------------- | -------------------------------------------------------------------- |
| `get_batch`                 | Fetch the status and details of a batch.                             |
| `get_file`                  | Fetch metadata for an uploaded file.                                 |
| `get_file_content`          | Fetch the raw content of an uploaded file (e.g. batch output JSONL). |
| `get_fine_tuning_job`       | Fetch the status and details of a fine-tuning job.                   |
| `get_model`                 | Fetch metadata for a single model by id.                             |
| `get_response`              | Fetch a stored model response by id.                                 |
| `get_vector_store`          | Fetch a vector store by id.                                          |
| `get_vector_store_file`     | Fetch the ingestion status of a file attached to a vector store.     |
| `list_batches`              | List the organization's batches.                                     |
| `list_files`                | List files uploaded to the organization.                             |
| `list_fine_tuning_events`   | List status events of a fine-tuning job.                             |
| `list_fine_tuning_jobs`     | List the organization's fine-tuning jobs.                            |
| `list_models`               | List the models available to the API key.                            |
| `list_response_input_items` | List the input items of a stored response.                           |
| `list_vector_store_files`   | List the files attached to a vector store.                           |
| `list_vector_stores`        | List the organization's vector stores.                               |
| `search_vector_store`       | Search a vector store for chunks relevant to a query.                |

## Write actions (17)

| Action                     | Description                                                               |
| -------------------------- | ------------------------------------------------------------------------- |
| `add_vector_store_file`    | Attach an already-uploaded file to a vector store for retrieval.          |
| `cancel_batch`             | Cancel an in-progress batch.                                              |
| `cancel_fine_tuning_job`   | Cancel a running fine-tuning job.                                         |
| `cancel_response`          | Cancel an in-progress background response.                                |
| `create_batch`             | Create an asynchronous batch of API requests from an uploaded JSONL file. |
| `create_chat_completion`   | Generate a model reply for a list of chat messages.                       |
| `create_embedding`         | Compute embedding vectors for one or more input texts.                    |
| `create_fine_tuning_job`   | Start a fine-tuning job from an uploaded training file.                   |
| `create_image`             | Generate an image from a text prompt.                                     |
| `create_moderation`        | Classify text against OpenAI content-safety categories.                   |
| `create_response`          | Generate a model response (Responses API) from text or structured input.  |
| `create_vector_store`      | Create a vector store, optionally seeded with uploaded files.             |
| `delete_file`              | Delete an uploaded file.                                                  |
| `delete_response`          | Delete a stored model response.                                           |
| `delete_vector_store`      | Delete a vector store.                                                    |
| `delete_vector_store_file` | Detach a file from a vector store (the file itself is not deleted).       |
| `update_vector_store`      | Update the name, expiry policy or metadata of a vector store.             |

## Permissions

Every action above must be explicitly granted to an agent before it can be used. See [Permissions](/integrations/permissions) for the grant model and [Security](/integrations/security) for how credentials are protected.