Lookalike Repo Finder

Find the top 20 most similar GitHub repos to any seed repo in under 30 seconds. Stop spending an afternoon clicking through GitHub topics, "you might also like" lists, and search results — get a ranked, current dataset.

What does Lookalike Repo Finder do?

You have a repo you love. Maybe it's the AI agent framework your team picked, or a library you want to evaluate alternatives for, or your own project (and you want to find competitors and adjacent communities). You want a list of repos that are actually similar — not the random results from a keyword search.

GitHub doesn't provide this. "Related repos" don't exist as a feature. You can manually pivot through topics, languages, and stars — but it takes hours and the result is biased toward whatever you noticed first.

This Actor:

1. Fetches the seed repo's metadata (topics, primary language, description)
2. Builds a candidate pool by searching GitHub for repos sharing the seed's topics + keywords + language
3. Scores every candidate on 4 dimensions: topic overlap (40 pts), language match (20 pts), recency (20 pts), popularity (20 pts) — max 100
4. Returns the top-N ranked dataset

The output is a single, current, decision-ready table.

Why use Lookalike Repo Finder?

• Competitor research: building an open-source library? Find what you're up against in 30 seconds
• Tech stack evaluation: comparing Next.js, Remix, Astro, Nuxt? Auto-discover the full landscape
• Investor / market research: see who else is operating in a category you're tracking
• Curation: build a "best of" list with objective ranking instead of vibes
• Job hunting: find every active company building products similar to one you admire

How to use Lookalike Repo Finder

1. Enter a seed repo in the form owner/repo (e.g. vercel/next.js)
2. Set Max results (default 20)
3. Optionally set Min stars (default 50) to filter out abandoned projects
4. Click Start
5. Download the dataset as JSON, CSV, or Excel

Input

• seedRepo (required) — A public GitHub repo to find lookalikes for. Example: vercel/next.js.
• maxResults (optional, default 20) — How many top lookalikes to return. Max 100.
• minStars (optional, default 50) — Filter out repos with fewer stars.

Output

Each row:

{
  "fullName": "remix-run/remix",
  "url": "https://github.com/remix-run/remix",
  "description": "Build Better Websites. Create modern, resilient user experiences with web fundamentals.",
  "stars": 30200,
  "language": "TypeScript",
  "topics": ["react", "remix", "ssr", "framework"],
  "lastPushed": "2026-05-18T21:14:33Z",
  "lookalikeScore": 88,
  "scoreBreakdown": {
    "topicOverlap": 30,
    "language": 20,
    "recency": 20,
    "popularity": 18
  }
}

You can download in JSON, HTML, CSV, or Excel from the Storage tab.

Quality scoring (4 dimensions, max 100 points)

Dimension	Max	What it measures
Topic overlap	40	Shared GitHub topics with the seed
Language match	20	Same primary language as seed
Recency	20	Pushed in last 30d=20, 90d=15, 180d=10, 365d=5
Popularity	20	Log-scaled stars (avoids over-rewarding mega-repos)

Pricing

• Per-search rate: $0.10 per seed repo (one full lookalike search)
• Bulk plan: $5 for 100 seed searches (great for org-wide audits)
• Custom: contact for white-label

Tips

• The seed repo must have GitHub topics for best results. If it has none, the Actor falls back to language + description keywords (less precise).
• For niche repos (under 100 stars), set minStars low (10 or 0) — otherwise the candidate pool will be empty.
• The Actor uses anonymous GitHub API (60 req/hour). For bulk runs, space them out or upgrade Apify proxy.

FAQ

Why are the results different from GitHub's "explore"?

GitHub's explore is editorial and platform-curated. This Actor is algorithmic and based on objective signals (topics, language, recency, popularity). The two complement each other.

Can this find private repos?

No. The Actor uses anonymous GitHub API. For private-repo lookalike search across your org, contact for a custom version with token support.

Why does the score sometimes seem off?

Topics are the strongest signal. If a repo doesn't use GitHub topics (~30% of repos don't), it'll score poorly even if it's a great match. This is a GitHub data quality issue, not an Actor bug.

Support

Issues / feature requests: open in the Issues tab on the Apify console.

Built by Ian Mu — github.com/ianymu — author of verify-before-stop, the Stop hook that blocks Claude Code session-end if files changed but no verification log exists.