Domain branding

The domain brand API collects a website’s observed logos, colors, fonts, descriptions and social links. An optional measured style guide adds typography, spacing and component properties. Every observation names its source pages, and conflicting observations remain visible. This is separate from the single-page branding format.

Get domain branding

Create an operation key once before a new request:

export URL2LLM_OPERATION_KEY="$(uuidgen)"

Save that value with your task. Keep it when rerunning the script for an identical retry; do not rerun the generation command. If uuidgen is unavailable, set your own unique value. For a different request, generate a new value. Node.js and Python require the saved key and use the shared HTTP client. Read retries before retrying a request whose result you did not receive.

curl -X POST https://api.url2llm.com/v1/brand \
  -H "Authorization: Bearer $URL2LLM_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: ${URL2LLM_OPERATION_KEY:?Set URL2LLM_OPERATION_KEY first}" \
  -d '{"domain":"example.com","includeStyleGuide":false}'

import { request } from './url2llm.mjs'

const operationKey = process.env.URL2LLM_OPERATION_KEY
if (!operationKey) throw new Error('Set URL2LLM_OPERATION_KEY to the saved operation key')
console.log('Operation key:', operationKey)
const result = await request('/v1/brand', {
  method: 'POST',
  operationKey,
  body: {
    "domain": "example.com",
    "includeStyleGuide": false
  },
})
console.log(result.data.id)

import os
from url2llm import request

operation_key = os.environ['URL2LLM_OPERATION_KEY']
if not operation_key:
    raise ValueError('Set URL2LLM_OPERATION_KEY to the saved operation key')
print('Operation key:', operation_key)
result = request(
    '/v1/brand',
    method='POST',
    operation_key=operation_key,
    body={'domain': 'example.com', 'includeStyleGuide': False},
)
print(result['data']['id'])

Use a key with scrape scope. Supply a hostname or HTTP(S) origin without a path, query or fragment. timeout defaults to 120,000 milliseconds and applies from the first execution across retries. Page selection is automatic.

Save the collection ID

Creation returns HTTP 202. Save data.id from the response:

{
  "success": true,
  "data": {
    "id": "COLLECTION_ID",
    "kind": "brand",
    "status": "queued",
    "total": 1,
    "completed": 0,
    "failed": 0,
    "pending": 1,
    "credits_used": 0,
    "reserved_credits": 10,
    "created_at": "2026-10-29T00:00:00Z",
    "updated_at": "2026-10-29T00:00:00Z",
    "warning": null
  },
  "request_id": "req_example"
}

Check progress

Set COLLECTION_ID to the returned ID. Use this ID for status, results, and cancellation with an active key from the same organization:

export COLLECTION_ID='YOUR_COLLECTION_ID'
curl "https://api.url2llm.com/v1/brand/$COLLECTION_ID/status" \
  -H "Authorization: Bearer $URL2LLM_API_KEY"

import { request } from './url2llm.mjs'

const collectionId = process.env.COLLECTION_ID
if (!collectionId) throw new Error('Set COLLECTION_ID to the returned ID')
const result = await request(`/v1/brand/${collectionId}/status`, { method: 'GET' })
console.log(result.data)

import os
from url2llm import request

collection_id = os.environ['COLLECTION_ID']
if not collection_id:
    raise ValueError('Set COLLECTION_ID to the returned ID')
result = request(f'/v1/brand/{collection_id}/status', method='GET')
print(result['data'])

A status response has the same fields as creation. For example:

{
  "success": true,
  "data": {
    "id": "COLLECTION_ID",
    "kind": "brand",
    "status": "running",
    "total": 1,
    "completed": 0,
    "failed": 0,
    "pending": 1,
    "credits_used": 0,
    "reserved_credits": 10,
    "created_at": "2026-10-29T00:00:00Z",
    "updated_at": "2026-10-29T00:00:00Z",
    "warning": null
  },
  "request_id": "req_example"
}

Repeat while data.status is queued or running. Terminal statuses are completed, failed, and cancelled. Do not create another extraction to check progress. See the collection polling example for a bounded polling loop.

Read the result

After completion, fetch the merged result:

curl "https://api.url2llm.com/v1/brand/$COLLECTION_ID" \
  -H "Authorization: Bearer $URL2LLM_API_KEY"

import { request } from './url2llm.mjs'

const collectionId = process.env.COLLECTION_ID
if (!collectionId) throw new Error('Set COLLECTION_ID to the returned ID')
const result = await request(`/v1/brand/${collectionId}`, { method: 'GET' })
console.log(result.data)

import os
from url2llm import request

collection_id = os.environ['COLLECTION_ID']
if not collection_id:
    raise ValueError('Set COLLECTION_ID to the returned ID')
result = request(f'/v1/brand/{collection_id}', method='GET')
print(result['data'])

{
  "success": true,
  "data": {
    "url": "https://example.com/",
    "facts": {
      "title": [{"value": "Example Domain", "sources": ["https://example.com/"]}]
    },
    "style_guide": null,
    "sources": ["https://example.com/"],
    "warning": null
  },
  "request_id": "req_example"
}

The response contains data.facts, optional data.style_guide, data.sources, and data.warning. Each observation has value and sources. Individual page content becomes available through GET /v1/brand/$COLLECTION_ID/results after the completed extraction is charged. It is withheld while running and after cancellation. Without includeStyleGuide, page results expose logos, colors and fonts but omit measured typography, spacing and component values.

Credits

Branding costs 10 credits. Set includeStyleGuide to true on the same endpoint to add the measured style guide for 5 more credits, totaling 15. The displayed price covers the complete extraction. If no useful evidence is found, or the extraction is cancelled before completion, the reserved credits are released.

Check missing facts

Missing facts remain absent. The output does not infer a company’s audience, tone, positioning or unsupported company details. The measured style guide reflects captured page styles; it does not reconstruct responsive layouts or all interactive states.

List previous extractions

GET /v1/brand lists extractions for the key’s organization. Read captured pages from GET /v1/brand/$COLLECTION_ID/results; use its integer next_cursor to paginate. Use the history pagination examples, replacing /v1/batch/scrape with /v1/brand. For captured pages, use the results loop with /v1/brand/$COLLECTION_ID/results after successful completion.

Cancel an extraction

Cancel pending work with the same collection ID:

curl -X DELETE "https://api.url2llm.com/v1/brand/$COLLECTION_ID" \
  -H "Authorization: Bearer $URL2LLM_API_KEY"

import { request } from './url2llm.mjs'

const collectionId = process.env.COLLECTION_ID
if (!collectionId) throw new Error('Set COLLECTION_ID to the returned ID')
const result = await request(`/v1/brand/${collectionId}`, { method: 'DELETE' })
console.log(result.data)

import os
from url2llm import request

collection_id = os.environ['COLLECTION_ID']
if not collection_id:
    raise ValueError('Set COLLECTION_ID to the returned ID')
result = request(f'/v1/brand/{collection_id}', method='DELETE')
print(result['data'])

Cancellation releases unused reserved credits. A completed extraction keeps its charge. Revoking the key that started the extraction also stops pending work.

Get started

API guides

Outputs

Reliability

Credits and billing

Domain branding

Get domain branding

Save the collection ID

Check progress

Read the result

Credits

Check missing facts

List previous extractions

Cancel an extraction

​Get domain branding

​Save the collection ID

​Check progress

​Read the result

​Credits

​Check missing facts

​List previous extractions

​Cancel an extraction

Get domain branding

Save the collection ID

Check progress

Read the result

Credits

Check missing facts

List previous extractions

Cancel an extraction