GitHub AI/ML Trending Scraper

GitHub AI/ML Trending Scraper extracts the trending repositories from github.com/trending — the same data GitHub shows on its Trending page, including the real star gain for the period (today, this week, or this month). It ships with a built-in AI / Machine Learning filter (pre-filled keywords and GitHub topics) so you get trending AI projects out of the box — but the filter is optional and fully customizable: turn it off to get every trending repo, or swap the keywords/topics to track any theme (web scraping, DevOps, game dev, anything). Run it on the Apify platform for scheduling, API access, proxy rotation, and integrations.

What does GitHub AI/ML Trending Scraper do?

It fetches one or more github.com/trending/<language> pages, parses every repository row, and returns structured data: repository name, description, language, total stars, forks, and stars gained in the selected period. By default it then keeps only repositories matching a pre-filled AI/ML keyword/topic list — but you can edit that list to filter by any theme, or disable filtering entirely to return the full trending list. Optionally, it enriches each repo with metadata from the GitHub API (topics, license, homepage, open issues, created/pushed dates).

Why use it?

• Track the AI/ML open-source landscape — see which models, agents, and libraries are breaking out today, this week, or this month.
• Tech newsletters & content — a ready-made, deduplicated list of trending AI projects.
• Investment / market research — star velocity is an early signal of developer traction.
• Competitive intelligence — monitor new AI tools as they gain momentum.

How to use it

1. Open the Actor and click Try for free / Start.
2. Pick a time range (Today / This week / This month).
3. Choose the programming languages to scan (default: python, jupyter-notebook).
4. Leave AI/ML repositories only enabled and pick a match mode (keywords, topics, or both).
5. Click Start and download the results as JSON, CSV, or Excel.

Input

Configure the run in the Input tab. Key fields:

Field	Description
since	Trending period: daily, weekly, or monthly.
languages	GitHub trending language slugs (e.g. python, jupyter-notebook, rust). Empty = all languages.
aiOnly	Keep only AI/ML repos, detected via a built-in topic + keyword dictionary (default true). Set false to return every trending repo.
filterTopics	Optional. Keep only repos having at least one of these GitHub topics (OR match). Empty = skip.
filterKeywords	Optional. Keep only repos whose name/description contains one of these words (OR match, whole-word). Empty = skip.
includeMetadata	Enrich each repo with GitHub API metadata.
maxRepos	Max results after filtering (sorted by star gain). 0 = no limit.
githubToken	Optional, secret. Raises the GitHub API rate limit from 60 to 5000 req/h for enrichment.
proxyConfiguration	Proxy used to fetch trending pages.

Example input:

{
  "since": "daily",
  "languages": ["python", "jupyter-notebook"],
  "aiOnly": true,
  "matchMode": "both",
  "maxRepos": 50
}

About the optional GitHub token

The token is never stored in the Actor's code. It is a per-run, secret input. It is only used to authenticate read-only calls to the public GitHub API so you get a higher rate limit. Without a token, enrichment still works but is limited to 60 requests/hour.

Output

Each dataset item represents one trending AI/ML repository. You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

{
  "rank": 1,
  "repository": "Shubhamsaboo/awesome-llm-apps",
  "owner": "Shubhamsaboo",
  "name": "awesome-llm-apps",
  "url": "https://github.com/Shubhamsaboo/awesome-llm-apps",
  "description": "100+ AI Agent & RAG apps you can actually run — clone, customize, ship.",
  "language": "Python",
  "stars": 119319,
  "forks": 17719,
  "starsInPeriod": 1006,
  "period": "daily",
  "topics": ["llm", "rag", "ai-agents"],
  "license": "Apache-2.0"
}

Data fields

Field	Description
rank	Position in the filtered, star-gain-sorted list.
repository	owner/name full repository slug.
url	Repository URL.
description	Repository description.
language	Primary programming language.
starsInPeriod	Stars gained in the selected period (today/week/month).
stars	Total stars.
forks	Total forks.
topics	GitHub topics (only when enrichment is enabled).
license, homepage, openIssues, watchers, createdAt, pushedAt	Extra metadata (only when enrichment is enabled).

Cost estimation

The trending pages are lightweight HTML — a run scraping a few languages costs a tiny fraction of a compute unit. Enrichment adds one GitHub API call per repository (free, rate-limited). For most users this comfortably fits within the Apify free tier.

Tips

• Add more languages (e.g. rust, typescript, c++) to widen coverage of AI/ML projects outside Python.
• Use matchMode: "both" for the widest AI/ML net; use topics for the most precise filtering.
• Provide a githubToken if you enable enrichment across many languages, to avoid the 60 req/h limit.
• Schedule the Actor daily to build a historical trend dataset.

FAQ and support

Is scraping GitHub Trending legal? This Actor only accesses publicly available pages and the public GitHub API. Use the data in compliance with GitHub's Terms of Service. It does not collect any personal or private data.

Why are some fields empty? topics and metadata fields are only populated when enrichment is enabled (includeMetadata, or matchMode using topics).

Found a bug or need a custom field? Open an issue in the Issues tab.