Extract page images

Use GET /web/scrape/images to find images referenced by a webpage. Results can include <img> sources, inline SVG, responsive sources, metadata images, favicons, CSS backgrounds, and video posters.

Prerequisites

Export an API key from the dashboard:

export CONTEXT_DEV_API_KEY="ctxt_secret_..."

Find images

The base request returns where each image was found, its source value, source type, and alt text when available.

curl --get https://api.context.dev/v1/web/scrape/images \
  --header "Authorization: Bearer $CONTEXT_DEV_API_KEY" \
  --data-urlencode "url=https://example.com"

import ContextDev from "context.dev";

const client = new ContextDev({
  apiKey: process.env.CONTEXT_DEV_API_KEY,
});

const page = await client.web.webScrapeImages({
  url: "https://example.com",
});

console.log(page.images);

import os
from context.dev import ContextDev

client = ContextDev(api_key=os.environ["CONTEXT_DEV_API_KEY"])

page = client.web.web_scrape_images(url="https://example.com")
print(page.images)

require "cgi/core"
require "context_dev"

client = ContextDev::Client.new(api_key: ENV.fetch("CONTEXT_DEV_API_KEY"))

page = client.web.web_scrape_images(url: "https://example.com")
puts page.images

package main

import (
    "context"
    "fmt"
    "os"

    contextdev "github.com/context-dot-dev/context-go-sdk/v2"
    "github.com/context-dot-dev/context-go-sdk/v2/option"
)

func main() {
    client := contextdev.NewClient(
        option.WithAPIKey(os.Getenv("CONTEXT_DEV_API_KEY")),
    )

    page, err := client.Web.WebScrapeImages(context.Background(), contextdev.WebWebScrapeImagesParams{
        URL: "https://example.com",
    })
    if err != nil {
        panic(err)
    }

    fmt.Println(page.Images)
}

<?php

require __DIR__.'/vendor/autoload.php';

use ContextDev\Client;

$client = new Client(apiKey: getenv('CONTEXT_DEV_API_KEY'));

$page = $client->web->webScrapeImages(url: 'https://example.com');
print_r($page->images);

The base request costs 1 credit, or 2 credits when it includes browser actions.

Enrich the images

Request enrichment only when you need to select, store, or categorize assets:

curl --get https://api.context.dev/v1/web/scrape/images \
  --header "Authorization: Bearer $CONTEXT_DEV_API_KEY" \
  --data-urlencode "url=https://example.com" \
  --data-urlencode "dedupe=true" \
  --data-urlencode "enrichment[resolution]=true" \
  --data-urlencode "enrichment[hostedUrl]=true" \
  --data-urlencode "enrichment[classification]=true"

const page = await client.web.webScrapeImages({
  url: "https://example.com",
  dedupe: true,
  enrichment: {
    resolution: true,
    hostedUrl: true,
    classification: true,
  },
});

page = client.web.web_scrape_images(
    url="https://example.com",
    dedupe=True,
    enrichment={
        "resolution": True,
        "hosted_url": True,
        "classification": True,
    },
)

require "cgi/core"
require "context_dev"

client = ContextDev::Client.new(api_key: ENV.fetch("CONTEXT_DEV_API_KEY"))

page = client.web.web_scrape_images(
  url: "https://example.com",
  dedupe: true,
  enrichment: {
    resolution: true,
    hosted_url: true,
    classification: true,
  },
)

puts page.images

package main

import (
    "context"
    "fmt"
    "os"

    contextdev "github.com/context-dot-dev/context-go-sdk/v2"
    "github.com/context-dot-dev/context-go-sdk/v2/option"
    "github.com/context-dot-dev/context-go-sdk/v2/packages/param"
)

func main() {
    client := contextdev.NewClient(
        option.WithAPIKey(os.Getenv("CONTEXT_DEV_API_KEY")),
    )

    page, err := client.Web.WebScrapeImages(context.Background(), contextdev.WebWebScrapeImagesParams{
        URL: "https://example.com",
        Dedupe: param.NewOpt(true),
        Enrichment: contextdev.WebWebScrapeImagesParamsEnrichment{
            Resolution: param.NewOpt(true),
            HostedURL: param.NewOpt(true),
            Classification: param.NewOpt(true),
        },
    })
    if err != nil {
        panic(err)
    }

    fmt.Println(page.Images)
}

<?php

require __DIR__.'/vendor/autoload.php';

use ContextDev\Client;

$client = new Client(apiKey: getenv('CONTEXT_DEV_API_KEY'));

$page = $client->web->webScrapeImages(
    url: 'https://example.com',
    dedupe: true,
    enrichment: [
        'resolution' => true,
        'hostedURL' => true,
        'classification' => true,
    ],
);

print_r($page->images);

Setting any enrichment flag makes the entire request cost 5 credits, including a request that also uses browser actions.

Read the result

{
  "success": true,
  "url": "https://example.com",
  "images": [
    {
      "src": "https://example.com/hero.png",
      "element": "img",
      "type": "url",
      "alt": "Example product dashboard",
      "enrichment": {
        "width": 1600,
        "height": 900,
        "mimetype": "image/png",
        "url": "https://media.brand.dev/example.png",
        "type": "photography"
      }
    }
  ],
  "cache_metadata": {
    "status": "miss",
    "age_ms": 0
  }
}

The src field can be an absolute URL, inline HTML such as SVG markup, or a base64 data URI. Branch on each item’s type before downloading or rendering it.

Choose enrichment

Requirement	Option	Returned fields
Compare image size	`enrichment.resolution=true`	`width`, `height`
Store a materialized copy	`enrichment.hostedUrl=true`	`url`, `mimetype`
Separate photos, logos, and icons	`enrichment.classification=true`	visual `type`
Collapse perceptual duplicates	`dedupe=true`	highest-resolution member of each duplicate group

Classification values are photography, illustration, logo, wordmark, icon, pattern, graphic, or other. Treat the value as a useful label, not proof that an asset is approved for a specific use.

Load more images

Use waitForMs for short delayed loads. Use browser actions when the page needs a click or scroll before images appear.

curl --get https://api.context.dev/v1/web/scrape/images \
  --header "Authorization: Bearer $CONTEXT_DEV_API_KEY" \
  --data-urlencode "url=https://example.com/gallery" \
  --data-urlencode 'actions=[{"do":"scroll","amount":"viewport","maxScrolls":5}]'

import ContextDev from "context.dev";

const client = new ContextDev({apiKey: process.env.CONTEXT_DEV_API_KEY});
const url = "https://example.com/gallery";

const page = await client.web.webScrapeImages({url}, {
  query: {
    url,
    actions: JSON.stringify([
      {
        "do": "scroll",
        "amount": "viewport",
        "maxScrolls": 5
      }
    ]),
  },
});

console.log(page.images);

import json
import os
from context.dev import ContextDev

client = ContextDev(api_key=os.environ["CONTEXT_DEV_API_KEY"])

page = client.web.web_scrape_images(
    url="https://example.com/gallery",
    extra_query={
        "actions": json.dumps([
          {
            "do": "scroll",
            "amount": "viewport",
            "maxScrolls": 5
          }
        ]),
    },
)

print(page.images)

require "cgi/core"
require "json"
require "context_dev"

client = ContextDev::Client.new(api_key: ENV.fetch("CONTEXT_DEV_API_KEY"))
actions = [{do: "scroll", amount: "viewport", maxScrolls: 5}]

page = client.web.web_scrape_images(
  url: "https://example.com/gallery",
  request_options: {extra_query: {"actions" => [JSON.generate(actions)]}},
)

puts page.images

package main

import (
    "context"
    "fmt"
    "os"

    contextdev "github.com/context-dot-dev/context-go-sdk/v2"
    "github.com/context-dot-dev/context-go-sdk/v2/option"
)

func main() {
    client := contextdev.NewClient(
        option.WithAPIKey(os.Getenv("CONTEXT_DEV_API_KEY")),
    )

    page, err := client.Web.WebScrapeImages(
        context.Background(),
        contextdev.WebWebScrapeImagesParams{URL: "https://example.com/gallery"},
        option.WithQuery("actions", `[{"do":"scroll","amount":"viewport","maxScrolls":5}]`),
    )
    if err != nil {
        panic(err)
    }

    fmt.Println(page.Images)
}

<?php

require __DIR__.'/vendor/autoload.php';

use ContextDev\Client;

$client = new Client(apiKey: getenv('CONTEXT_DEV_API_KEY'));

$page = $client->web->webScrapeImages(
    url: 'https://example.com/gallery',
    actions: [
        ['do' => 'scroll', 'amount' => 'viewport', 'maxScrolls' => 5],
    ],
);

print_r($page->images);

The TypeScript, Python, Ruby, and Go examples send actions as a JSON query value to preserve the array in SDK 2.14.0.

Production behavior

Results describe assets referenced by the rendered page. They do not guarantee that every URL remains downloadable later.
alt can be null; do not use it as the only source of an accessible label in your product.
maxAgeMs defaults to one day, accepts up to 30 days, and can be set to 0 for a fresh scrape.
Forwarding custom headers bypasses cache reads and writes.
Use the returned cache_metadata to distinguish a hit, miss, or honored ZDR request.

Next steps

Images API reference

Review the complete image schema and error responses.

Get started

Web data

Brand data

Automation

Production

Prerequisites

Find images

Enrich the images

Read the result

Choose enrichment

Load more images

Production behavior

Next steps

Images API reference

​Prerequisites

​Find images

​Enrich the images

​Read the result

​Choose enrichment

​Load more images

​Production behavior

​Next steps

Images API reference

Prerequisites

Find images

Enrich the images

Read the result

Choose enrichment

Load more images

Production behavior

Next steps