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

# Quickstart

> Create a Gabin account and an API key, then call the API and connect the MCP server.

**1. Create an account**

Go to [gabin.ai/compte/connexion](https://gabin.ai/compte/connexion) and sign in with GitHub or Google. The account is created at the first sign-in, with your verified email.

**2. Create an API key**

Open [gabin.ai/compte/api](https://gabin.ai/compte/api) and create a key. It starts with `gab_live_` and is shown **once**: copy it and store it as a secret.

```bash theme={null}
export GABIN_API_KEY="gab_live_…"
```

**3. Make your first call**

Find a company by name:

```bash curl theme={null}
curl "https://api.gabin.ai/v1/companies?q=electricite%20de%20france&limit=3" \
  -H "Authorization: Bearer $GABIN_API_KEY"
```

```javascript JavaScript theme={null}
const res = await fetch(
  "https://api.gabin.ai/v1/companies?q=electricite%20de%20france&limit=3",
  { headers: { Authorization: `Bearer ${process.env.GABIN_API_KEY}` } },
);
const { data, meta } = await res.json();
console.log(data.map((c) => `${c.siren} ${c.name}`), meta.total);
```

```python Python theme={null}
import os, requests

r = requests.get(
    "https://api.gabin.ai/v1/companies",
    params={"q": "electricite de france", "limit": 3},
    headers={"Authorization": f"Bearer {os.environ['GABIN_API_KEY']}"},
)
r.raise_for_status()
print([(c["siren"], c["name"]) for c in r.json()["data"]])
```

**4. Follow the links**

Take a SIREN from the result and open the company, then its buildings:

```bash theme={null}
curl "https://api.gabin.ai/v1/companies/552081317" -H "Authorization: Bearer $GABIN_API_KEY"
curl "https://api.gabin.ai/v1/companies/552081317/buildings" -H "Authorization: Bearer $GABIN_API_KEY"
```

**5. Check your usage**

`GET /v1/me` returns your account, your remaining rate limit and daily quota, and today's usage.

```bash theme={null}
curl "https://api.gabin.ai/v1/me" -H "Authorization: Bearer $GABIN_API_KEY"
```

```json theme={null}
{
  "data": {
    "account_id": "gRHviIMFzmqtukFO",
    "credential": { "type": "api_key", "id": "ga7_YGhxgiwrJieG" },
    "plan": "free",
    "rate_limit": { "limit": 60, "remaining": 53, "reset_seconds": 48 },
    "quota": { "limit": 5000, "remaining": 4993, "reset_seconds": 37428 },
    "usage_today": { "requests": 6, "units": 6 }
  },
  "meta": {}
}
```

## Use it from an AI assistant

Add `https://mcp.gabin.ai/mcp` as a connector in Claude or ChatGPT, or register it in Claude Code or Cursor with your key. See [Connect the MCP server](https://docs.gabin.ai/mcp/connect.md).

## Next steps

> [card] [Limits](https://docs.gabin.ai/limits.md)
>
> Rate limit, daily quota, request weights.

> [card] [Conventions](https://docs.gabin.ai/conventions.md)
>
> Response envelope, pagination, formats.

> [card] [Buildings guide](https://docs.gabin.ai/guides/buildings.md)
>
> From an address to the companies in the building.

> [card] [Errors](https://docs.gabin.ai/errors.md)
>
> Every error code and what to do.


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