GitHub Repository Scraper

Search GitHub repositories or look up specific repos through the official GitHub REST API and get clean rows: stars, forks, topics, language, license, dates, homepage, plus optional README text and language breakdown. Use any GitHub search qualifiers, or ask for newly created, fast-growing repos. No proxy, no login, fast, and $0.80 per 1,000 repos.

What you can do with it

• Find trending repos: topic:llm with "created within 30 days" and sort by stars lists what is taking off right now.
• Build lead lists and market maps: every repo in a topic or language, with owner, homepage and license, for outreach or competitor tracking.
• Track open-source projects: stars, forks, open issues and last push date for a watchlist of owner/name repos.
• Feed RAG and research pipelines: repo metadata plus README text in one dataset.

Input

Field	What it does
queries	owner/name, a github.com URL, or a search (topic:web-scraping language:python stars:>500, or plain keywords).
sort	stars (default), forks, updated or best-match.
createdWithinDays / pushedWithinDays	Trending (new) or still-maintained repos.
minStars, language	Quick filters added to your search.
includeForks	Forks are skipped by default.
maxPerQuery	Repos per search, up to the GitHub limit of 1,000 (default 100).
includeLanguages, includeReadme	Extra fields, one extra GitHub request per repo.
keywords	Keep only repos whose metadata contains one of these.
maxItems	Hard cap on results (and cost).

Example:

{ "queries": ["topic:llm", "psf/requests"], "createdWithinDays": 30, "minStars": 50, "maxPerQuery": 50 }

Output

One item per repo:

{ "fullName": "psf/requests", "owner": "psf", "url": "https://github.com/psf/requests", "description": "A simple, yet elegant, HTTP library.",
  "stars": 54460, "forks": 9800, "language": "Python", "topics": ["client", "http"], "license": "Apache-2.0",
  "createdAt": "2011-02-13T18:38:17Z", "pushedAt": "2026-10-01T12:00:00Z", "homepage": "https://requests.readthedocs.io",
  "languages": {"Python": 123456}, "readme": "..." }

Pricing

Pay per result: $0.80 per 1,000 repos (repo-scraped), with lower prices on higher Apify plans. Filters and caps apply before you are charged. A run that finds nothing fails instead of silently returning an empty dataset.

Limits and notes

• Uses the official GitHub API. GitHub search returns at most 1,000 results per query and allows about 30 searches per minute; the actor waits out short rate limits and stops cleanly (keeping what it found) on long ones.
• README and language options cost one extra API request per repo, so they are slower on big runs.
• Public repository data only. Follow GitHub terms and applicable privacy law if you use owner data for outreach.

Source code

This actor is open source (MIT): github.com/tidyharvest/open-actors/tree/main/github_repos_scraper. Issues and pull requests are welcome there.