# Hugging Face

> Let the Coder search models, datasets, Spaces, and papers on the Hugging Face Hub, and read one repository. Read only. Coming soon.

Status: Coming soon

Hugging Face hosts models, datasets, Spaces, and papers. Once you connect it, the Coder can search the Hub and read one repository, including the private repositories a read token can see.

> [!SOON]
> The Hugging Face integration is Coming soon. This page describes how it will work. The integrations open at launch are GitHub, GitLab, and Bitbucket.

## Which assistants use it

| Assistant | Status |
| --- | --- |
| [Coder](/docs/personas/coder) | Coming soon |

## What assistants can do

Hugging Face connects through the Hub's own hosted connector, and it is read only.

| Ability | How it runs |
| --- | --- |
| Search models, datasets, Spaces, and papers | Reads on its own |
| Read a repository's details | Reads on its own |
| Read the account the token belongs to | Reads on its own |

## Setup (when it opens)

1. In Hugging Face, open **Settings**, then **Access Tokens**, and create a fine-grained token that can read the repositories this assistant should see. A read token is enough. It starts with `hf_`.
2. Open [Integrations](https://stellarfirm.ai/app/integrations) in StellarFirm and pick Hugging Face.
3. Paste the token, switch on the live option, and press Connect.

## Permissions

- Nothing here waits for an Approve because nothing here changes data.
- Public Hub search can answer without a token. The token is what lets the Coder see private repositories and which account is signed in.
- Revoke the token in Hugging Face whenever you want to cut access.

## Prompts to try

```prompt title="Find a model"
Coder, search the Hub for a text model that can read a long document, and tell me which one to try first, with a link to its page.
```

---

Source: https://stellarfirm.ai/docs/integrations/huggingface