> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.usescout.sh/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.usescout.sh/_mcp/server.

# Pricing

Scout bills by credits. A credit is a single unit of work — a search, a
page read, an agent step. Every response tells you exactly what the call
cost so you can budget and compare.

## Plans

| Plan        | Monthly | Annual (per mo) | Credits / month | Overage             | Best for                                                       |
| ----------- | ------- | --------------- | --------------- | ------------------- | -------------------------------------------------------------- |
| **Free**    | \$0     | \$0             | 1,500           | —                   | Trying Scout out, prototypes, side projects. No card required. |
| **Builder** | \$29    | \$23            | 5,500           | \$9 / 1,000 credits | Single-app production traffic.                                 |
| **Scale**   | \$99    | \$79            | 25,500          | \$6 / 1,000 credits | High-volume agents and teams shipping multiple products.       |

Choose monthly or annual billing in the [Scout dashboard](https://platform.usescout.sh).
Annual saves about 20% over month-to-month billing.

Need more than Scale? [Contact us](mailto:support@usescout.sh) and we will
quote a custom plan based on your traffic shape.

## What a credit costs

Per-call costs are deterministic and shown in the `credits` field on every
response.

| Endpoint                           | Credits                      | Notes                                                 |
| ---------------------------------- | ---------------------------- | ----------------------------------------------------- |
| `/v1/search` (sync)                | 1                            | One credit per query in the request.                  |
| `/v1/search` (deep / async)        | 1 + 1 per page fetched       | Charged when the run completes.                       |
| `/v1/extract`                      | 2 per URL                    | Includes the parent and any auto-discovered subpages. |
| `/v1/task` (research)              | 1 per agent step             | Steps include planner calls, fetches, and synth.      |
| `/v1/chat/completions`             | 1 + 1 if `web_search: true`  | Plus one credit per web fetch the model triggers.     |
| `/v1/findall`                      | 2 per candidate row returned |                                                       |
| `/v1/monitors` (per scheduled run) | Same as the underlying query | Webhook deliveries do not cost extra.                 |

Opt-in features — objective rerank, content extraction, summaries — are
billed as separate line items so you can see the breakdown.

## Reading `usage[]`

Every response carries a per-SKU breakdown under `usage`:

```json
{
  "search_id": "search_4f5e...",
  "results": [...],
  "credits": 3,
  "reranked": true,
  "usage": [
    { "name": "sku_search", "count": 1 },
    { "name": "sku_objective_rerank", "count": 1 },
    { "name": "sku_contents_highlights", "count": 5 }
  ]
}
```

| SKU                       | Triggered by                            | What it covers                                              |
| ------------------------- | --------------------------------------- | ----------------------------------------------------------- |
| `sku_search`              | Any `/v1/search` call                   | One Google SERP fetch per query.                            |
| `sku_objective_rerank`    | `objective` on `/v1/search`             | One LLM call that re-ranks the merged results. Cached.      |
| `sku_contents_text`       | `advanced_settings.contents.text`       | One page fetch and clean per result.                        |
| `sku_contents_highlights` | `advanced_settings.contents.highlights` | One LLM call per result that pulls focused excerpts.        |
| `sku_contents_summary`    | `advanced_settings.contents.summary`    | One LLM call per result that writes the summary.            |
| `sku_subpage_rank`        | `objective` on `/v1/extract`            | One LLM call that picks the most relevant subpages. Cached. |
| `sku_extract`             | Every fetched URL                       | Includes parent and subpages.                               |

If a SKU does not appear in `usage`, it was not used and you were not
billed for it.

## Free-tier limits

Free accounts are capped at 1,500 credits per month. The cap resets on
the first of each month, in UTC. Once you exhaust the cap, requests
return `429 Too Many Requests` until the next reset or until you upgrade.

See [Rate limits](/rate-limits) for the in-flight request rules that
apply to every plan.

## Overage and rollover

* **Builder and Scale** plans allow overage. Calls past your monthly
  allocation are billed at the per-credit rate of your plan.
* Unused credits do not roll over.
* Cancellation takes effect at the end of the current billing period —
  no pro-rated refunds.

## See it in your dashboard

The [Settings → Usage](https://platform.usescout.sh) page shows live
credit consumption, per-endpoint breakdown, and a 30-day chart. Set a
soft alert there if you want a heads-up before you hit your cap.