BrightData n8n integrationAutomate BrightData with n8n.

Pushes a finished marketplace snapshot into storage you control, instead of leaving it to be downloaded later.

• Snapshot ID: the snapshot to operate on, often {{ $json.snapshot_id }} carried over from the filter step.
• Filename Template: the naming pattern for the delivered file, placeholders included, paired with File Extension set to JSON, JSONL or CSV.
• Deliver Type: picks the destination, from Amazon S3 or Google Cloud Storage to SFTP, Snowflake, Microsoft Azure, Aliyun Object Storage Service, Google Cloud PubSub or a plain Webhook. Each one then asks for its own fields, such as Bucket and AWS Access Key, or Host and Username.

Narrows a marketplace dataset down to the records you want and turns the result into a snapshot you can read afterwards.

• Dataset: the dataset to work on, selected from the list rather than typed.
• Filter Type: Single Filter for one condition, Group Filters when several conditions combine.
• Field Name, Operator and Field Value: the condition itself, with operators covering Equals, Not Equals, Greater Than, Includes, In and their negatives.
• Records Limit: caps how many records end up in the snapshot.

Describes one dataset rather than returning its records: what it holds and how it is shaped, which is what you need before writing a filter against it.

• Dataset: the only field to fill, picked from the list of datasets available on the account.

Returns the actual records held in a marketplace snapshot, in batches when the volume justifies it.

• Snapshot ID: the snapshot to read, typically the one produced by the filter step.
• Batch Size: how many records come back per response, with Part selecting which batch, numbered from 1.
• Format: JSON, JSONL or CSV, and Compress to get the response gzipped.

Gives the description of a marketplace snapshot without pulling any of its records, which keeps the step cheap when all you want is context.

• Snapshot ID: the single required field, the identifier of the snapshot to inspect.

Cuts a marketplace snapshot into parts and tells you how it was divided, so the download can be spread over several executions.

• Snapshot ID: the snapshot to split, nothing else to configure.

Returns the marketplace datasets the account can reach, with no parameters at all to fill in.

Returns the snapshot identifiers already created on a given marketplace dataset, so a workflow can pick one up instead of generating another.

• Dataset: the dataset whose snapshots you want listed.
• Status: filters on the snapshot state, from Building, Collecting and Digesting through to Ready, Failed or Canceled.
• View: the identifier of a view used to narrow the list further.

Same destination options as its marketplace counterpart, applied to a snapshot your own scraping run produced, plus a way to be told when the transfer is done.

• Notify: a URL called once the delivery finishes, which an n8n Webhook can receive to continue the workflow.
• Snapshot ID, Filename Template and File Extension: what to ship, under what name, in JSON, JSONL or CSV.
• Deliver Type: S3, Google Cloud Storage, Azure, SFTP, Snowflake, PubSub, Aliyun or Webhook, each with its own required fields.

Brings the scraped records back into the workflow as items, rather than sending them to external storage.

• Snapshot ID: the batch to download, usually the identifier returned when the collection was started.
• Format: JSON or CSV, required either way.
• Split Into Parts: switch it on for large snapshots, then use Batch Size and Part to pull them in chunks.

Searches your scraping snapshots over a time window instead of listing everything ever produced.

• Dataset: which dataset the search runs against.
• From Date and To Date: the window, both required, in ISO 8601 format.
• Status: restricts the answer to one state, Ready and Failed being the two that matter most in practice.
• Limit: caps the number of results, with Skip moving past the first ones.

Answers a single question about a running scrape: where is it now. Collection is asynchronous, so this is what turns a snapshot ID into a decision.

• Snapshot ID: the extraction being watched, the only field to set.

Runs one extraction and gives you the structured result directly, without the snapshot round trip that batch collection requires.

• Dataset: the dataset that defines what gets extracted from the page.
• URLs: the addresses to scrape, passed as JSON, which can come from an earlier node.
• Include Errors: required, and worth leaving on so a failed page is reported rather than silently dropped.
• Format: JSON or CSV.

Starts a scraping job over many addresses at once and returns an identifier, not the data. The rows arrive later.

• Dataset: the extraction template applied to every address in the batch.
• URLs: the list to collect, as JSON.
• Endpoint: where the collected data gets sent once it exists.
• Notify: a URL called when the collection finishes, which removes the need to keep checking.

Sends a query to Google and returns the results page as data, which is the fastest way to get search visibility into a workflow.

• Zone: the Bright Data zone the request goes through.
• Country: where the search is performed from, which changes what Google answers.
• Search Query: the query string itself.
• Page: 1 for the first results page, 2 for the next, and so on.

Behaves like a request against one page, routed through a zone, and hands back whatever that page returns.

• URL and Method: the target and the verb, GET, POST, PUT, PATCH, DELETE or HEAD.
• Zone and Country: the route the request takes.
• Format: Raw or JSON, and Data Format set to Markdown when the answer feeds a language model, which forces the format back to Raw.