GitHub Repos & Trending Scraper – Stars, Issues & Releases

Scrape GitHub repositories, search results, GitHub trending repos and developers, issues, pull requests and releases: stars, forks, watchers, topics, license, languages, latest release, last commit, contributors count. Official API + github.com/trending, optional token.

Price: $1 per 1,000 results. Works without a GitHub account, and an optional free token raises the API limit to 5,000 requests per hour.

What data can you get from GitHub?

Mode	What you get
🔥 Trending	github.com/trending by language, period (today / this week / this month) and spoken language: rank, repo, description, language, stars, forks, stars gained in the period, "built by" logins. Trending developers: rank, login, profile URL and their popular repository.
🔎 Search repositories	GitHub repository search with keywords and any qualifier (topic:, org:, pushed:>…), language and minimum-stars filters, sorted by best match, stars, forks or recently updated. Up to 1,000 results per query.
📦 Repository details	For a list of owner/repo or URLs: stars, forks, watchers, open issues, topics, license, homepage, default branch, archived/fork flags, created/updated/pushed dates. With standard details you also get the languages breakdown (bytes and %), latest release, last commit date and contributors count. Full adds a README excerpt.
🐞 Issues & pull requests	Issues and/or PRs filtered by state (open/closed/all), labels and "updated since": number, title, state, labels, assignees, milestone, comments, reactions, created/updated/closed/merged dates, author login and a short body excerpt.
🚀 Releases	Tag, name, publish date, pre-release flag, author login, assets with sizes and download counts, total downloads and a release-notes excerpt.

Privacy by design: the Actor never collects e-mail addresses or other personal data. People are identified only by their public GitHub login and profile URL, and e-mail addresses inside issue bodies, README and release notes are replaced with [email removed].

Use cases

• Developer-tool market research. Find and rank the repositories in your niche by stars, forks and activity.
• Trend spotting. Run a daily schedule on trending repos for Python, Rust, TypeScript or any other language and spot rising projects early.
• Competitive intelligence. Track competitors' stars, releases, release download counts and issue activity over time.
• Open-source due diligence. Check license, last commit date, contributors count and open issues before you adopt a dependency.
• Lead lists for DevRel and investors. Build lists of fast-growing projects. Only public logins and URLs are included, never e-mails.
• AI agents and RAG. Give an LLM structured, current GitHub data such as repo stats and README excerpts.

How to use the GitHub Scraper

1. Click Try for free.
2. Pick What to scrape (mode). Trending works with no other input.
3. Fill in the field for your mode: Search query for search, or Repositories for details, issues and releases.
4. Optional: add a GitHub token for larger runs (see below).
5. Click Start, then download your data as JSON, CSV or Excel, or use the API.

Do I need a GitHub token?

No, but large runs go much faster with one.

	Without token	With token
Trending	unlimited (HTML page)	unlimited
REST API (details, issues, releases)	60 requests/hour per IP, shared with others	5,000 requests/hour
Search API	10 requests/minute	30 requests/minute

A fine-grained personal access token with no permissions is enough, because it only reads public data. Create one at https://github.com/settings/personal-access-tokens. The field is a secret input and is stored encrypted.

What one result costs in API requests: a search request returns up to 100 repositories, an issues or releases request up to 100 rows, and repository details take 1 request (basic), 5 (standard) or 6 (full) per repository.

Rate limits are handled for you. The Actor reads GitHub's x-ratelimit-* headers. If the limit resets within Max wait for rate limit (default 90 s), it waits and continues. Otherwise it stops cleanly, keeps everything scraped so far and adds a free RATE_LIMITED status row that tells you when the limit resets.

Input example

{
    "mode": "repoDetails",
    "repositories": ["apify/crawlee", "https://github.com/sindresorhus/awesome"],
    "detailLevel": "full",
    "maxItems": 50
}

Other examples:

{ "mode": "trending", "language": "python", "trendingPeriod": "weekly" }
{ "mode": "trending", "trendingType": "developers", "language": "rust" }
{ "mode": "searchRepositories", "searchQuery": "web scraping", "language": "python", "minStars": 500, "sort": "stars", "maxItems": 100 }
{ "mode": "issues", "repositories": ["apify/crawlee"], "issueType": "issues", "issueState": "all", "labels": ["bug"], "since": "90 days" }
{ "mode": "releases", "repositories": ["microsoft/vscode"], "includePrereleases": false, "maxItemsPerRepository": 10 }

Running with no input ({}) returns today's trending repositories.

Output

Every row has a type: repository, trending-repository, trending-developer, issue, pull-request, release, or the free rows status and error. Below is real output, shortened.

Repository details (detailLevel: full)

{
    "type": "repository",
    "fullName": "apify/crawlee",
    "owner": "apify",
    "ownerType": "Organization",
    "ownerUrl": "https://github.com/apify",
    "url": "https://github.com/apify/crawlee",
    "description": "Crawlee—A web scraping and browser automation library for Node.js to build reliable crawlers. ...",
    "homepage": "https://crawlee.dev",
    "stars": 26028,
    "forks": 1688,
    "watchers": 136,
    "openIssues": 140,
    "primaryLanguage": "TypeScript",
    "topics": ["apify", "automation", "crawler", "crawling", "headless", "headless-chrome", "..."],
    "license": "Apache-2.0",
    "isArchived": false,
    "createdAt": "2016-08-26T18:35:03Z",
    "pushedAt": "2026-10-07T18:39:46Z",
    "languages": [
        { "language": "TypeScript", "bytes": 4295265, "percent": 62.2 },
        { "language": "MDX", "bytes": 2133590, "percent": 30.9 },
        { "language": "JavaScript", "bytes": 371816, "percent": 5.4 }
    ],
    "latestRelease": { "tagName": "v3.18.2", "name": "v3.18.2", "publishedAt": "2026-09-29T09:58:35Z", "url": "https://github.com/apify/crawlee/releases/tag/v3.18.2" },
    "lastCommitDate": "2026-10-06T14:13:47Z",
    "contributorsCount": 152,
    "readmeExcerpt": "A web scraping and browser automation library Crawlee covers your crawling and scraping end-to-end and helps you build reliable scrapers. ...",
    "detailLevel": "full",
    "scrapedAt": "2026-10-07T19:06:31.912Z"
}

Trending repository

{
    "type": "trending-repository",
    "rank": 1,
    "fullName": "Panniantong/Agent-Reach",
    "url": "https://github.com/Panniantong/Agent-Reach",
    "description": "Give your AI agent eyes to see the entire internet. ...",
    "language": "Python",
    "stars": 93110,
    "forks": 8158,
    "starsInPeriod": 6454,
    "builtBy": ["Panniantong", "claude", "nyxst4ck", "kadaliao", "hymansun"],
    "period": "weekly",
    "trendingLanguage": "python"
}

Trending developer

{ "type": "trending-developer", "rank": 1, "login": "penso", "url": "https://github.com/penso", "popularRepo": "penso/herdr-gpui", "popularRepoUrl": "https://github.com/penso/herdr-gpui", "popularRepoDescription": "Native client for Herdr on macOS, Linux, and Windows, built with Rust and GPUI. ...", "period": "weekly", "trendingLanguage": "rust" }

Issue

{
    "type": "issue",
    "repository": "apify/crawlee",
    "number": 2815,
    "title": "bug: `setValue` and `useState` functions fail for large objects",
    "state": "closed",
    "stateReason": "completed",
    "url": "https://github.com/apify/crawlee/issues/2815",
    "authorLogin": "lhotanok",
    "authorUrl": "https://github.com/lhotanok",
    "labels": ["bug", "t-tooling"],
    "assignees": ["janbuchar"],
    "milestone": "4.1",
    "comments": 2,
    "createdAt": "2025-01-16T13:38:56Z",
    "closedAt": "2026-09-24T11:37:05Z",
    "bodyExcerpt": "Which package is this bug report for? ..."
}

Release

{ "type": "release", "repository": "microsoft/vscode", "tagName": "1.141.0", "name": "1.141.0", "url": "https://github.com/microsoft/vscode/releases/tag/1.141.0", "isPrerelease": false, "publishedAt": "2026-10-07T11:01:16Z", "authorLogin": "alexr00", "assetsCount": 0, "totalDownloads": 0, "assets": [], "bodyExcerpt": "https://code.visualstudio.com/updates/v1_141" }

Status (free)

{ "type": "status", "statusCode": "RATE_LIMITED", "message": "GitHub core API rate limit reached (unauthenticated: 60 requests/hour, search 10/minute). It resets at 2026-10-07T20:06:31.000Z ... Add a free GitHub personal access token ...", "rateLimitResetAt": "2026-10-07T20:06:31.000Z" }

Status codes: RATE_LIMITED, NO_RESULTS, NO_RELEASES. Error codes: INVALID_INPUT, NOT_FOUND, INVALID_QUERY, SCRAPE_FAILED. Status and error rows are never charged.

A few notes on the fields. openIssues is GitHub's own count, which includes open pull requests. watchers is the real subscriber count and is only filled in repository details, because search results don't include it. contributorsCount is an estimate: GitHub counts anonymous contributors and returns null for repositories that are too large to list. Renamed repositories are followed automatically, so fullName shows the current name.

The dataset has ready-made views: Repositories, Trending, Issues & pull requests, Releases and Status & errors.

Pricing

Pay-per-event: you pay only for results, and platform usage is included.

Event	Price	When it is charged
item	$0.001 per result ($1 / 1,000)	Each repository, trending repo or developer, issue, pull request or release saved
Status / error rows	Free	Rate-limit notices, empty results, invalid or missing repositories

Job	Cost
Today's trending repositories (≈ 13–25 rows)	≈ $0.02
100 search results	$0.10
Details for 50 repositories	$0.05
1,000 issues	$1.00

Set Max results and Apify's Max cost per run. The Actor stops exactly at your budget.

Use with AI agents (MCP)

This Actor works as a tool for AI agents through the Apify MCP server (Claude, Cursor, VS Code, OpenAI Agents and others). Add github-scraper to your MCP server tools and the agent can call it with inputs like these:

{ "mode": "repoDetails", "repositories": ["vercel/next.js", "facebook/react"], "detailLevel": "standard" }

{ "mode": "trending", "language": "typescript", "trendingPeriod": "daily", "maxItems": 10 }

The output is flat JSON with stable field names, so agents can compare repositories, summarize releases or triage issues directly.

FAQ

Is it legal to scrape GitHub? The Actor reads only public data: the official public REST API and the public trending page. It does not log in, does not access private repositories and does not collect e-mail addresses or other personal contact data. You are responsible for using the data in line with GitHub's Terms of Service and its Acceptable Use Policies, including the rules on information usage. Don't use the data for spam or unsolicited contact, and keep the limits of GDPR and similar laws in mind.

Why did my run stop with RATE_LIMITED? Without a token, GitHub allows only 60 API requests per hour per IP address, and Apify's IPs are shared. Add a free token, raise Max wait for rate limit, or run again after the time shown in the status row. Results already scraped are kept, and you only pay for those.

Why does search stop at 1,000 results? That is a GitHub Search API limit. Split your query with qualifiers (for example stars:100..500, created:2025-01-01..2025-06-30 or language:go) to cover more.

Can I get stargazer lists or follower e-mails? No. You get stargazer counts, not lists of people, and never e-mail addresses. This is deliberate.

Can I run it on a schedule? Yes. Use Apify Schedules, for example daily trending for your languages or weekly release tracking for your competitors.

Which language names work for trending? Use the names GitHub shows: python, typescript, c++, c#, jupyter notebook, go, rust and so on. Leave the field empty for all languages.

Related actors

• Coming soon: GitLab Scraper
• Coming soon: npm & PyPI Package Stats
• Coming soon: Hacker News Scraper

Support & feedback

Found a bug or need a field? Open an issue in the Issues tab. We usually respond within a day.