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

# Quickstart

> Create a brand, add a prompt, and read its results in a few calls.

This walks through the full loop: create a brand, add a prompt, optionally set a
location, and read results. Every call needs your
[API key](/authentication) — set it once:

```bash theme={null}
export AGL_KEY="agl_live_your_key_here"
```

## 1. Check your plan

Your plan sets how many brands and prompts you can have, and whether brands are
scanned daily.

```bash theme={null}
curl https://agentled.co/api/v1/plan \
  -H "Authorization: Bearer $AGL_KEY"
```

```json theme={null}
{
  "plan": "pro",
  "features": { "brands": 1, "prompts": 50, "frequency": "daily", "models": ["chatgpt"] }
}
```

## 2. Create a brand

Pass a website. The name, description, and logo are enriched automatically.

```bash theme={null}
curl -X POST https://agentled.co/api/v1/brands \
  -H "Authorization: Bearer $AGL_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "website": "example.com" }'
```

```json theme={null}
{ "brand": { "id": "b1a7…", "domain": "example.com", "name": "Example", "status": "active", "location": { "mode": "worldwide", "country": null, "city": null, "label": null }, "…": "…" } }
```

Save the brand `id` — the next calls use it. Creating the same domain again just
returns the existing brand (it won't duplicate or re-enrich).

<Warning>
  A new brand has **no prompts**, and a brand with no active prompts is **never
  scanned**. Add at least one prompt in the next step.
</Warning>

## 3. Add a prompt

Prompts are the questions asked about the brand each scan.

```bash theme={null}
curl -X POST https://agentled.co/api/v1/brands/BRAND_ID/prompts \
  -H "Authorization: Bearer $AGL_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "text": "What are the best example tools?" }'
```

```json theme={null}
{ "prompt": { "id": "p3c9…", "text": "What are the best example tools?", "active": true, "…": "…" }, "usage": { "used": 1, "limit": 50 } }
```

## 4. (Optional) Set a location

By default a brand is measured worldwide. To measure a specific market, discover
valid values and set them. See [Concepts → Location](/concepts#location).

```bash theme={null}
# Discover valid values
curl https://agentled.co/api/v1/geo/countries -H "Authorization: Bearer $AGL_KEY"
curl https://agentled.co/api/v1/geo/countries/US/cities -H "Authorization: Bearer $AGL_KEY"

# Set the brand to a city
curl -X PATCH https://agentled.co/api/v1/brands/BRAND_ID \
  -H "Authorization: Bearer $AGL_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "location": { "mode": "city", "country": "US", "city": "Austin" } }'
```

## 5. Wait for a scan, then read results

On a paid plan the brand is scanned **daily** once it has active prompts. After a
scan completes, read its metrics and history:

```bash theme={null}
# Visibility metrics over the last 30 days
curl "https://agentled.co/api/v1/brands/BRAND_ID/metrics?days=30" \
  -H "Authorization: Bearer $AGL_KEY"

# Scan history
curl https://agentled.co/api/v1/brands/BRAND_ID/scans \
  -H "Authorization: Bearer $AGL_KEY"

# Every answer a prompt received
curl https://agentled.co/api/v1/brands/BRAND_ID/prompts/PROMPT_ID/answers \
  -H "Authorization: Bearer $AGL_KEY"
```

<Note>
  Scans aren't triggered through the API yet. Brands are picked up by the daily
  sweep, which runs on **paid** plans only — so a brand created via the API on a
  free plan can be configured but won't produce results until it's on a paid
  plan.
</Note>

## Next steps

* [Concepts](/concepts) — brands, prompts, location, and plan limits in depth.
* [Errors & pagination](/errors) — the error shape and how to page long lists.
* [API reference](/introduction) — every endpoint, generated from the spec.