# Read a company's website and extract pricing with Bytemine's API

Bytemine's Live Company Intelligence API reads a company's own web pages at request time, not a stored profile, and returns either the page's content or structured fields, such as the plans on a pricing page. A page read costs $0.01 and a schema extraction $0.02. Send `POST https://agents.bytemine.ai/zcj/c5w5ntc1gx61/functions/v1/api-gateway/company/intel/page` or `.../company/intel/extract` with the page `url` and pay the request's 402 price quote from a wallet over x402 or MPP; no account or API key is needed.

## At a glance

- **Endpoints:** `POST https://agents.bytemine.ai/zcj/c5w5ntc1gx61/functions/v1/api-gateway/company/intel/page` to read one page, `.../company/intel/extract` for structured fields
- **Body:** `{"url": "<page URL>"}` for a read. `{"url": "<page URL>", "schema": "pricing"}` for an extraction.
- **Price:** $0.01 per page read, $0.02 per extraction
- **Payment:** each request answers with a 402 price quote, paid from a wallet over x402 or MPP, or drawn from prepaid Credits.
- **Response:** a read returns `url`, `success`, `markdown`, `metadata`, `links` and `tech_stack`. An extraction returns the fields of the schema you chose.
- **Limits:** one URL and one schema per extraction. A bulk read takes up to 100 URLs.
- **Full purchase flow:** [the buyer skill](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/SKILL.md)

## Read a pricing page and extract its plans

1. Read the page live:

   ```bash
   curl -i -X POST https://agents.bytemine.ai/zcj/c5w5ntc1gx61/functions/v1/api-gateway/company/intel/page \
     -H 'Content-Type: application/json' \
     -d '{"url": "https://www.shopify.com/pricing"}'
   ```

   The first answer is `402 Payment Required`, with the amount in `payment.amountUsd`.

2. Pay the quote with an x402 or MPP client, then retry the same request with the proof attached. The [buyer skill](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/SKILL.md) has the steps. After payment the response is JSON of this shape:

   ```json
   {"url": "<string>", "success": "<boolean>", "markdown": "<page content>", "metadata": {}, "links": ["<string>"], "tech_stack": {}}
   ```

   The example request above paid $0.01.

3. To get fields instead of page text, extract the same page with the `pricing` schema:

   ```bash
   curl -i -X POST https://agents.bytemine.ai/zcj/c5w5ntc1gx61/functions/v1/api-gateway/company/intel/extract \
     -H 'Content-Type: application/json' \
     -d '{"url": "https://www.shopify.com/pricing", "schema": "pricing"}'
   ```

   Pay the 402 and retry as in step 2. The response carries `success`, `schema` and these pricing fields:

   ```json
   {"plans": [{"name": "<string>", "price": "<string>", "billing_period": "<string>", "target_audience": "<string>", "features": ["<string>"]}], "currency": "<string>", "pricing_model": "<string>", "has_free_tier": "<boolean>", "enterprise_available": "<boolean>"}
   ```

   The example extraction of that page paid $0.02.

4. Check the `zc-billing` header on each paid response. It reconciles the amount you authorized (`authorizedUsd`) with the amount you were charged (`chargedUsd`).

## Choose an extraction schema

| What you need from the page | `schema` |
| --- | --- |
| Plans, prices, billing periods, free tier | `pricing` |
| What the company is and does | `company_profile` |
| Technologies the page reveals | `tech_stack` |
| Links to the company's social profiles | `social_links` |
| Brand elements | `branding` |
| All of these in one extraction | `full_intelligence` |

Each extraction takes one schema, and `schema` must be one of the pre-built values above.

## Other live operations

Each operation below is a `POST` to its path on `https://agents.bytemine.ai/zcj/c5w5ntc1gx61`.

| Path | What it does | Body | Listed at |
| --- | --- | --- | --- |
| `/functions/v1/api-gateway/company/intel/pages` | Reads up to 100 URLs in one call and returns `results` | `urls` | the page rate |
| `/functions/v1/api-gateway/company/intel/sitemap` | Maps the site's structure and returns `links` and `total` | `url`, optional `limit` (default 100), `search` | the page rate |
| `/functions/v1/api-gateway/company/intel/scan/start` | Starts a multi-page scan and returns `job_id` and `status` | `url`, optional `max_depth` (default 3), `max_pages` (default 50) | the page rate |
| `/functions/v1/api-gateway/company/intel/scan/continue` | Resumes pending pages and returns `pages_crawled` and `pages_pending` | `job_id` | the page rate |
| `/functions/v1/api-gateway/company/intel/report` | Builds one report from the company's highest-value pages, with `extractions` and `pages_analyzed` | `url` | the page rate |
| `/functions/v1/api-gateway/company/intel/snapshot` | Saves a point-in-time copy of a page and returns `snapshot_id` and `created_at` | `url` | the page rate |
| `/functions/v1/api-gateway/company/intel/changes` | Compares a page's live content with its saved snapshot and returns `added`, `removed` and `change_summary` | `url` | the page rate |

To track a scan, send its `job_id` to `/functions/v1/api-gateway/company/intel/scan/status`. To read its results, send it to `/functions/v1/api-gateway/company/intel/scan/results`, which pages them with `page` and `per_page` (default 50). The [service's API reference](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/services/live-company-intelligence/openapi.json) has every request and response schema.

## Costs and boundaries

- A page read is a fixed $0.01. Sitemaps, scans, reports, snapshots and change checks bill $0.01 per page they process, so their total depends on the site; `max_pages` caps a scan. For these calls the 402 can be a ceiling, the most the call can cost, with only the pages actually processed charged.
- Card pays a pay-as-you-go call only at $0.50 or more. A single read or extraction is below that, so it goes over x402 or MPP, or draws on prepaid Credits.
- Credits start at $0.50, or $0.50 by card; buying them needs an agent credential from [auth.md](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/auth.md).

## Errors

| Error | Meaning | Action |
| --- | --- | --- |
| `402` with no `reason` | The per-call price quote | Pay it and retry the same request, or stop |
| `settlement_failed` | The signed payment did not settle; `settlementReason` says why | Fix the wallet, then send a fresh request for a new quote |
| `payment_consumed` | The payment was already spent | Check whether the earlier request delivered before paying again |
| `usage_exhausted` | The Credits balance is used up | Top up with `POST https://agents.bytemine.ai/zcj/c5w5ntc1gx61/extend` and `{"amountUsd": "..."}` |
| `invalid_request` (400) | The body is malformed | Fix the JSON |

After a paid call is forwarded, the status and body come from Bytemine's API, errors included. [The error reference](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/errors.md) lists every code.

## Frequently asked questions

### Does Bytemine read the live website or a cached copy?

The live website. These operations read the company's own pages at request time, so a pricing page read today returns what that page says today.

### Can I extract pricing from many pages in one call?

No. An extraction takes one `url` and one `schema`, so each page is its own call at $0.02. The `pages` operation reads up to 100 URLs in one call, but it returns page content, not schema fields.

### How do I monitor a company website for changes?

Save the page with `snapshot`. Later, send the same `url` to `changes`. It compares the live page with the saved snapshot and returns `added`, `removed` and a `change_summary`. Each bills at the page rate.

### Do I need a Bytemine account or API key?

No. Each call is paid at the request through its 402 quote. A bearer credential is needed only to buy Credits.

### How is a full-site scan or report billed?

Per page processed, at $0.01 a page, so the total depends on how many pages the call reads. `max_pages` caps a scan, and each call's 402 states the most it can cost before you pay.

## Keep reading

- [Buyer skill](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/SKILL.md): the full purchase flow
- [Live Company Intelligence API reference](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/services/live-company-intelligence/openapi.json): every operation's schema
- [How to pay](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/payment.md): x402, MPP, card and Credits
- [Error reference](https://agents.bytemine.ai/zcj/c5w5ntc1gx61/errors.md): every error code and what to do

---

Updated 2026-10-16. All Bytemine guides: https://agents.bytemine.ai/guides.md. Storefront: https://agents.bytemine.ai/zcj/c5w5ntc1gx61/llms.txt.