Scout is in active development. Public launch: 1 December 2026. Until then, try Scout live and read the docs — public signup and API keys are not open yet.

Scout in development · launch 1 Dec 2026. Try it live · docs

Website Inspection API

Getscreenshotsas API.

Scout opens pages in a real browser and returns raw data: screenshots, DOM extracts, a11y facts, and journeys. No SEO scores — just reproducible evidence.

Try it live Open docs

Real browser execution Structured raw outputs OpenAPI-first

https://example.com/pricing

desktop · 1280×720

mobile

POST /v1/screenshot/json · live

Call it your way

HTTP, TypeScript SDK, or CLI — same API underneath.

TypeScript SDK

Typed Node.js client for screenshots, asserts, journeys, and jobs.

@macrocontent/scout

Install

pnpm add @macrocontent/scout

Example

import { Scout } from '@macrocontent/scout'

const scout = new Scout({ apiKey: process.env.SCOUT_API_KEY! })
await scout.screenshot({ url: 'https://example.com', mode: 'viewport' })

CLI

Terminal client for scripts and CI — built on the SDK.

@macrocontent/scout-cli

Install

pnpm add -g @macrocontent/scout-cli

Screenshot

scout screenshot https://example.com -o shot.jpg

Assert

scout assert https://example.com/pricing \
  --rule '#buy|visible' \
  --rule '.price|contains:29'

One API, three jobs

Capture evidence, verify expectations, then automate and monitor.

Capture and extract

Pull screenshots, DOM, and raw browser state from a real page load.

POST /v1/screenshot

Screenshot API

Capture viewport or full-page images as visual evidence for QA and monitoring.

Use when: When you need a visual artifact attached to a CI run, alert, or regression review.

example.com/pricingshots[2]

jpeg · 1280×720 + 390×844base64 ✓

POST /v1/extract

DOM Extraction

Structured extracts from a real browser — returned as JSON for pipelines.

Use when: When agents or scripts need page content as data, not as a rendered image.

a.primary → cta

.plan → count

"title": "Pricing",

POST /v1/extract

XHR Capture

Record matching XHR and fetch bodies from a real page load — pricing APIs, carts, search.

Use when: When the data you need lives in network JSON, not in the rendered DOM.

captured_xhr[1]fetch

GET /api/pricing

{ "amount": 2990 }

capture_xhr: */api/pricing*

POST /v1/extract

Adaptive Selectors

Relocate CSS selectors when markup drifts; generate paths; collect similar card lists.

Use when: When monitoring a live page whose classes or list markup change between deploys.

selector_relocationsrelocated

from a.btn-primary

to a.checkout-cta#buy

adaptive + match_text: Buy

POST /v1/extract

HTTP Fetch Mode

Load a static document without JavaScript — cheaper and faster when you do not need a full browser.

Use when: When the HTML is already in the first response and you do not need client-rendered DOM.

JavaScript disabled

2 credits · static document

POST /v1/checks/*

Browser Facts API

Retrieve raw cookies, console, network, JSON-LD, and metadata.

Use when: When you need browser state and telemetry — not a pass/fail rule.

networkcookiesjson-ldconsole

json_ld.types: ["Product", "Offer"] · top_slowest: true

Test and verify

Run focused checks, multi-rule assertions, accessibility, and performance.

POST /v1/checks/a11y-snapshot

Accessibility Facts

Structured a11y snapshots from a live page — facts you can store and compare.

Use when: When you want accessibility evidence as data for audits, not a single score.

a11y_snapshotinventory

landmarks: main · nav · footer

POST /v1/checks/*

Page Checks API

Run one focused verification such as visibility, security, or tech facts.

Use when: When a single targeted check is enough — one concern, one endpoint call.

/v1/checks/visibilityPASS

overall_pass: truein_viewport: truenot_covered: true

securitya11y · axetech

POST /v1/assert

Assert API

Evaluate several explicit expectations against one shared page load.

Use when: When CI needs many rules on the same visit — cheaper and more consistent than N checks.

failed_count: 0passed

checks: visible · exists

a[href*=signup]

count_min ≥ 3 · actual 3

POST /v1/performance

Performance & Sandbox

Timing metrics and isolated HTML sandboxes for bundle and render checks.

Use when: When you need load timing or a sandboxed HTML run without crawling production.

/v1/performance

screenshot_base64 ✓

Automate and monitor

Multi-step journeys, scheduled jobs, and crawl-driven URL discovery.

POST /v1/journey

Browser Journeys

Multi-step interact → assert flows with optional debug output.

Use when: When a single page load is not enough — login, click, fill, then verify.

interact · click

a[href*=signup]

+ debug.screenshot

Website Monitoring

Scheduled checks with signed webhooks for alerting pipelines.

Use when: When the same verification should run on a schedule and notify on failure.

job · */15running

assert · h1PASS

job_id: 7f3a…

X-Scout-Signature

Crawl API

DNS, robots, and link discovery for audits, monitoring, and URL queues.

Use when: When you need to discover URLs before screenshotting, asserting, or monitoring them.

discovered_urls10 pages

● https://example.com/

├─ /pricing

└─ /blog +7

dns A/AAAArobots allowcrawl 10

Use cases

Scout as an evidence layer for QA, agent workflows, and monitoring.

Automated QA

Assertions and screenshots as release gates in CI.

AI Agents

Browser evidence as an API layer for agent workflows.

Website Monitoring

Jobs, webhooks, and alerting for ongoing quality checks.

Self-hosted

Run Scout in your infrastructure

Inspect private applications, internal environments, and sensitive workflows with annual licensing and no usage credits.

Explore self-hosted Compare cloud and self-hosted

Private networks, staging, and intranet — without cloud egress of page content.
Same API surface as cloud — license by workers, not per-request credits.
Docs: self-hosted install

Ready for real browser evidence?

Public keys open on 1 December 2026. Until then, try Scout live and read the docs.

Try it live Scout docs