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

# Files

> The free dataset and the shared Parquet file of every LinkedIn profile fetched through iBlinked.

```bash theme={null}
curl https://idle.datacircle.dev/files/ -H "Authorization: Token $DATACIRCLE_API_KEY"
```

Each file has a `rank` (its tier), `row_count`, `size_bytes`, `sha256`, `created_at`, and `allowed`, which says whether this workspace may download it. `row_count` and `sha256` are `null` when we have not measured that file (today: `sha256` on both US datasets, `row_count` on `large`):

| Tier | What | Who |
| - | - | - |
| `free` | the 10M+ US B2B leads dataset: a 2.8 GB zip of two Parquet files, 11.1M people and 1.75M companies | everyone |
| `large` | the 50M+ US B2B dataset: an 11.9 GB zip of `person_us.parquet` and `company_us.parquet` | 3 people you invited signed up, or your workspace bought \$50 of credit |
| `sample` | the first 1,000 rows of today's profile file | everyone |
| `full` | every public LinkedIn profile anyone fetched through iBlinked, as one Parquet file (Up2Data calls are not in it yet) | 3 people you invited signed up, or your workspace bought \$50 of credit |

`full` is a new file every day, with the `sample` cut from it. Each one is a full snapshot of everything so far, so `GET /files/` lists only the newest file of each rank.

```bash theme={null}
curl -X POST https://idle.datacircle.dev/files/$FILE_ID/download-link/ \
  -H "Authorization: Token $DATACIRCLE_API_KEY"
```

Returns a download `url` valid for one hour, or `403` if the file is not unlocked for this workspace.

## Invites

```bash theme={null}
curl https://idle.datacircle.dev/me/invites/ -H "Authorization: Token $DATACIRCLE_API_KEY"
```

Returns your `invite_link`, how many people you invited have `signed_up` (verified their work email), and how many are `needed` (3). Send the link; once 3 have signed up, `large` and `full` unlock for you.

The file only holds public LinkedIn profile fields. It never contains who fetched what, emails of Datacircle users, costs or any credential.

```python theme={null}
import pandas as pd
profiles = pd.read_parquet("linkedin_profiles_full.parquet")
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.