Top 200 ReposEXPLORER
Browse agents, categories and pages

Discover the Projects and AI Work That Matter

Top 200 Repos follows the most-starred public repositories on GitHub with exact, recorded star counts, and connects them to source-linked examples of what people actually do with AI agents. Every number comes from a stored snapshot, and every use case links to where it came from.

What You Can Explore

Each area links to a working page. The examples are read from the current data, so they change as the catalog changes.

How Rankings Work

The ranking is a record of what GitHub reported, not an opinion about which projects are best.

  1. Official GitHub Search

    A scheduled job asks the GitHub Search API for public repositories with more than 1,000 stars, sorted by stars, and reads the first two pages of 100 results. The exact query is:

    stars:>1000 is:public fork:true
  2. Exact Counts, Recorded

    Each star count is the exact integer the API returned at retrieval time. The full result is stored as a snapshot with its timestamp.

  3. History from Snapshots Only

    Ranks, history and growth are calculated only between stored snapshots. Nothing is interpolated, back-filled or estimated.

  4. Known Limits

    GitHub Search indexing, candidate scope and pagination limit global completeness. Forks and archived repositories are included as GitHub returns them, and counts do not change between snapshots.

Popularity is not a certification. A high star count says nothing about a project’s security, quality, maintenance or fitness for your purpose. Read the source before you depend on it.

How Use Cases Are Evaluated

A use case is only as good as its source, so each entry keeps the claim and the evidence apart.

  1. Original Sources

    Every entry links to a public original: project documentation, a first-person account, a vendor post, news, a paper or a discussion. The source type is shown on the entry.

  2. Claims Versus Verified Results

    What a source says happened is labelled as reported. An entry is marked independently verified only after a recorded independent check.

  3. Editorial Review

    Submissions go to a review queue. Nothing is published before review, and published entries are written in our own words.

  4. Attribution and Corrections

    Credit goes to the original author where known. Each use-case page has a Report a Problem form, and corrections can also be sent through the contact page.

Outcome Labels

Every use case carries one of these labels. The count is the number of entries in the library with that label today.

  • Reported by source45 entries

    The source says this happened. Top 200 Repos has not reproduced it.

  • Independently verified0 entries

    Reproduced or confirmed by an independent check.

  • Planned, not done22 entries

    The source describes an intention, not a finished result.

  • Outcome unclear0 entries

    The source does not say how it turned out.

Freshness and Transparency

These times are read from the server for this page view, in UTC.

GitHub ranking snapshot
Use-case library updated
Video guide list updated
Next scheduled GitHub check

The ranking is refreshed by a daily job scheduled for 19:00 UTC. If a run fails, the last good snapshot stays in place and the failure is shown in the source status below rather than hidden. Growth figures always compare two snapshots that were actually recorded; where history is missing, the site shows a not-enough-history message instead of a trend.

Source Status

Data sourcesLast syncedLast checked
GitHubLive11 Nov 2026, 17:19 UTC11 Nov 2026, 17:19 UTC
YouTubeNot connectednever11 Nov 2026, 17:19 UTC
RedditNot connectednever11 Nov 2026, 17:19 UTC
Hacker NewsLive11 Nov 2026, 17:20 UTC11 Nov 2026, 17:20 UTC
XNot connectednevernever

Built for Curious People

There is no single right way to use the catalog. A few starting points:

  • Developers

    Find well-known tools in your language and see how others pair them with coding agents.

    Filter the rankings

  • Researchers

    Compare recorded star histories and follow each claim back to its source.

    See observed growth

  • Founders

    Scan which agents are applied to which jobs, with sources to read in full.

    Browse jobs

  • Students

    Start from a topic and work outward to the repositories and guides behind it.

    Explore topics

  • Teams and Businesses

    Read documented outcomes before trying an agent, and notice what is only planned or unclear.

    Read use cases

  • Non-Coders

    Video guides and plain-language summaries explain what a project does without reading code.

    Watch video guides

How to Participate

No account is needed for any of these.

  • Browse and Save

    Save use cases and repositories on this device and come back to them later.

    Open saved items

  • Share

    Every repository and use case has a Share button that copies its permanent link.

    Find something to share

  • Suggest a Use Case

    Send a public source and a short description. It is reviewed before anything is published.

    Suggest a use case

  • Report Something Wrong

    Use Report a Problem on a use case, or describe the error on the contact page.

    How to report an error

  • Follow New Use Cases

    Subscribe to the RSS feed in any feed reader.

    Read the RSS guide

  • Weekly Digest

    Sign-ups are recorded now. Email sending is not switched on yet, and the first email will ask you to confirm.

    Join the digest list

Independence and Trust

  • Top 200 Repos is an independent project. It is not affiliated with, endorsed by or sponsored by GitHub or any AI agent vendor. Product and project names belong to their owners.
  • Sponsorship never changes rankings, star counts, outcome labels or the order of use cases. A sponsor placement is labelled “Sponsored” and kept apart from editorial content.
  • Links to repositories, sources and videos lead to third-party sites with their own terms and privacy practices.

Start Exploring

Pick a starting point. Everything links back to its source.