GitHub Weekly Trending Repos & Org Repo Finder

GitHub Trending, Developer & Organization Repository Scraper

Track what GitHub is amplifying right now, then go deeper: scrape the Trending page for repositories and developers, search every public repository, list an organization's or user's repos, or pull complete details for repositories you already care about. Every request is cookieless and works without a GitHub account or token.

What you can collect

Six modes share one output shape:

Mode	What it does
trending	The actual github.com/trending list, filtered by daily/weekly/monthly period, programming language and spoken language
developers	The github.com/trending/developers list: developer, avatar, profile URL and the repository that put them on the list
search	Keyword search across public repository names, descriptions and READMEs, with server-side language, star and push-date filters
orgRepos	Every public repository owned by an organization, paginated up to 1,000 results
userRepos	Every public repository owned by a user, paginated up to 1,000 results
repoDetails	Full metadata for a list of owner/repo paths or GitHub URLs, with optional language breakdown, top contributors and README markdown

Output fields

Repository records use these fields (empty values are omitted, never null):

• Identity — repoId, repoName, owner, ownerType, ownerUrl, ownerAvatarUrl, fullName
• Visibility — visibility, defaultBranch, isForked, isArchived, isTemplate, allowForking
• Popularity — stars, forks, watchers, subscribers, openIssues, networkCount
• Trending — trendingRank, trendingPeriod, starsGained, starsGrowthPercent, builtBy[]
• Content — description, homepage, topics[], language, languages (byte breakdown), languagesCount
• Dates — createdAt, updatedAt, pushedAt
• License — license, licenseKey, licenseSpdxId, licenseUrl
• Links — repoUrl, cloneUrl, sshUrl, stargazersUrl, forksUrl, contributorsUrl, releasesUrl
• Detail extras — contributors[], readme
• Metadata — recordType (github_repo or github_developer), scrapedAt

Developer records carry username, displayName, avatarUrl, profileUrl, popularRepoName, popularRepoFullName, popularRepoDescription, popularRepoUrl, trendingRank, trendingPeriod, recordType and scrapedAt. Set includeDeveloperProfiles to add bio, followers, following, publicRepos, company, location, blogUrl, twitterUsername, profileCreatedAt and profileUpdatedAt.

Input

Field	Default	Notes
mode	trending	One of the six modes above
since	weekly	daily, weekly or monthly for trending/developers
searchQuery	machine learning	Free-text query, GitHub search qualifiers allowed
language	all	Programming language dropdown; customLanguage overrides it
spokenLanguage	all	README language filter for trending
customSpokenLanguage	—	Free-text ISO code that overrides the spoken-language dropdown
org / user	openai / torvalds	Organization or user login (URLs accepted)
repoUrls	[]	owner/repo paths or GitHub URLs for repoDetails
sortBy	stars	stars, forks, updated, help-wanted-issues or best-match
minStars	0	Minimum star count; 0 disables
pushedWithinDays	0	Only repos pushed within N days; 0 disables
aiOnly	false	Keep AI/ML-related repos based on topics and descriptions
filterTopics	[]	Keep repos with at least one of these topics
includeForks	false	Include forked repositories
includeDeveloperProfiles	false	Profile bio and stats (developers)
includeLanguages	true	Language byte breakdown (repoDetails)
includeContributors	false	Top 10 contributors (repoDetails)
includeReadme	false	README markdown (repoDetails)
maxItems	20	1–1,000

Examples

This week's trending Rust repositories:

{ "mode": "trending", "language": "Rust", "since": "weekly", "maxItems": 20 }

Trending developers this month:

{ "mode": "developers", "since": "monthly", "maxItems": 25 }

Search for actively maintained vector databases:

{
  "mode": "search",
  "searchQuery": "vector database",
  "language": "Python",
  "minStars": 500,
  "pushedWithinDays": 90,
  "sortBy": "stars",
  "maxItems": 50
}

Every public repo in an organization, newest activity first:

{ "mode": "orgRepos", "org": "vercel", "sortBy": "updated", "maxItems": 100 }

Full details plus contributors for three repositories:

{
  "mode": "repoDetails",
  "repoUrls": ["openai/whisper", "https://github.com/rust-lang/rust", "pallets/flask"],
  "includeContributors": true,
  "includeReadme": true,
  "maxItems": 3
}

How it works

• Trending is scraped from the public GitHub Trending HTML page, so it reflects the same ordering GitHub shows. The page exposes about 25 repositories per period/language; filters are applied after parsing.
• Search, org and user modes use GitHub's public search API. The search UI and API both cap a single query at 1,000 results; narrow with language, stars, dates or topics to cover more.
• Repository details use the public repository endpoint and, only when you enable them, the languages, contributors and README endpoints.
• If the unauthenticated API limit is reached, the actor falls back to GitHub's public HTML pages where possible and says so in the run status. Trending mode does not depend on the API.
• Every field is written only when the source has a real value; empty values and empty lists are omitted at every depth.

No credentials, no proxy setup

The actor uses public GitHub pages and endpoints only. You do not need a GitHub token, cookie, login or proxy configuration. Unauthenticated GitHub API allowances are shared per IP, which is enough for normal runs; very large searches may pause briefly or fall back to HTML.

FAQ

Is this the same list as github.com/trending? Yes for trending and developers: those modes parse the Trending pages directly, including the selected period and language.

Why did trending return fewer than maxItems? GitHub Trending publishes roughly 25 repositories or developers per period/language combination. Switch language or period to see a different slice.

Why is my search capped at 1,000 records? That is GitHub's search API limit for any single query. Split the request by language, topic, star range or keyword to collect more.

Why are some fields missing on trending records? Trending-page fields are always present when GitHub shows them. Deeper metadata (repository ID, topics, license, dates) comes from the public repository endpoint; if that endpoint is rate-limited, the actor keeps the trending-page fields instead of inventing values.

Are forks included? No by default. Set includeForks to true for search, org and user modes.

Can I use saved inputs from the previous version? Yes. Existing mode, searchQuery, language, org, sortBy and maxItems inputs keep working. The actor also recognizes common legacy aliases such as query, maxRepos, scrapeType and resultsPerQuery in code, without cluttering the input form.

How does aiOnly work? A repository counts as AI-related when its GitHub topics match an AI/ML list or its name/description contains AI/ML phrases such as “large language model”, “computer vision”, “RAG” or “fine-tuning”. Use filterTopics for exact topic control.

Is this affiliated with GitHub? No. This is an independent third-party actor that reads publicly available GitHub data. It is not endorsed by or affiliated with GitHub, Inc.

Data source

Public GitHub Trending pages (github.com/trending), the public GitHub REST API (api.github.com), and raw.githubusercontent.com for optional README downloads. No authentication is used.