Google Patent Details Scraper

Google Patent Details Scraper fetches known public Google Patents detail pages by publication number or canonical Google Patents URL.

Quick start

{
  "publicationNumbers": ["US9014905B1"],
  "maxItems": 1,
  "detailLevel": "metadata"
}

Use the docs/publication/INPUT-EXAMPLES.json for publication numbers, canonical URLs and full detail.

What you can enter

• publicationNumbers for known public publications;
• patentUrls for exact HTTPS Google Patents detail pages;
• language, detailLevel, maxItems and includePdfLink for bounded detail extraction.

Keyword search, application-number search, filters, sort, page controls and search-capacity controls are not part of this release. The Actor does not silently turn those inputs into an empty result; the active schema rejects them and the runtime records a contract failure if a schema-bypassing caller supplies them.

What you receive

Each saved Dataset row has one publication number and its canonical source URL. Common fields include the title, abstract, inventors, assignees, dates, CPC and IPC classes, source status, citations, simple-family data and an optional PDF link. Use detailLevel: "full" for claims, description and other bounded fields from the same patent page. PDF and image assets are never downloaded.

The default key-value store contains RUN_SUMMARY. It records direct-detail requests, saved rows, truncation, technical outcomes and every input outcome.

Safe limits

• up to 100 publication numbers or exact patent URLs;
• up to 100 saved unique publications across the whole run;
• one bounded direct detail request for each accepted publication.

Result statuses

PATENT_SAVED and PARTIAL are Dataset rows. NOT_FOUND, INVALID_INPUT, BLOCKED, RATE_LIMITED, TRANSIENT_ERROR and PARSER_DRIFT are recorded in RUN_SUMMARY when no complete row is saved. A source block is never returned as an empty success.

Source and legal note

Results come from public Google Patents detail pages. Some fields can be missing or translated. Source status, expiration dates and events are not legal advice or a legal conclusion. Verify important facts with the relevant patent office.

This independent tool is not affiliated with or endorsed by Google LLC. It uses direct HTTPS for detail pages, does not log in, solve challenges or use another patent database, and never downloads patent documents.

Need help?

Include the Apify run ID, input JSON without secrets and the related RUN_SUMMARY entry. See the docs/publication/SUPPORT.md for status meanings and safe troubleshooting.