Use the public Context.dev API in n8n workflows to search and scrape the live web, extract structured data, monitor websites, process batches, find company news, parse documents, and enrich brand or person data.
The npm package remains named n8n-nodes-branddev so existing installations and workflows continue to upgrade normally. New nodes use the Context.dev name and the current production API at https://api.context.dev/v1.
In n8n, open Settings → Community Nodes, select Install, and enter:
n8n-nodes-branddev
See n8n's community node installation guide for self-hosted installation options.
- Create a Context.dev account at context.dev.
- Copy an API key from the Context.dev dashboard.
- In n8n, create a Context.dev API credential and paste the key.
The credential check calls a read-only account-limits endpoint and does not create work or consume web-data credits.
The current node version tracks every operation in the public API reference.
| Resource | Operations |
|---|---|
| Web Scraping | Scrape Markdown, scrape HTML, scrape images, crawl a sitemap, crawl a website, search the web, and take screenshots |
| Web Extraction | Extract structured data, styleguides, fonts, NAICS/SIC classifications, and one or many products |
| Brand Intelligence | Retrieve a brand by domain, name, email, ticker, direct URL, or transaction; search brands |
| News | Search current company news by name, domain, ticker, or ISIN |
| Parsing | Convert binary PDF, Office, image, text, code, and data files into Markdown |
| People | Enrich a person from email, social profiles, name, company, education, or location clues |
| Monitors | Create, list, retrieve, update, delete, run, and inspect monitors, runs, changes, usage, limits, and webhook secrets |
| Batch | Submit, list, retrieve, cancel, delete, and read results for scrape or crawl batches |
| Webhooks | List, retrieve, inspect attempts for, and retry webhook deliveries |
| Utility | Prefetch brand or styleguide data |
Current operations accept the same JSON objects documented by the API. Every operation starts with a valid minimal example that can be edited or replaced with an n8n expression.
For example, News → Search Company News starts with:
{
"searchBy": {
"type": "entity",
"entity": {
"type": "domain",
"domain": "context.dev"
}
},
"limit": 10
}To get a timestamped YouTube transcript, choose Web Scraping → Scrape Markdown and pass the video URL:
{
"url": "https://www.youtube.com/watch?v=VIDEO_ID"
}For path-based operations, IDs are separate required fields and are URL-encoded automatically. Destructive operations require an explicit confirmation toggle. Batch submission and webhook retry also expose an optional idempotency key.
The Parsing → Parse File operation reads binary data produced by a previous n8n node. Set Input Binary Field to that field's name (usually data). The node sends the bytes directly, infers the extension from n8n's binary metadata when possible, and enforces the API's 25 MiB limit before upload.
- New Context.dev nodes default to version 2 and use only current public endpoints.
- Existing version 1 Brand.dev nodes retain their saved resources, operations, and request shapes.
- The internal node and credential identifiers are unchanged, so upgrading does not orphan saved workflows or credentials.
- All requests use
https://api.context.dev/v1; the oldapi.brand.devhostname is no longer emitted.
Use Node.js 24 for the current n8n development server.
npm ci
npm run check
npm run devnpm run check runs n8n's strict Cloud-compatible lint rules, unit and execution tests, a live contract comparison against Context.dev's published OpenAPI document, a production build, and a production-dependency audit.