Footer

Your go-to destination for discovering trending open-source projects and uncovering the insights that matter

Legal

  • Terms and conditions
  • Privacy policy

More

  • Pricing
  • Advertise
  • onedump
  • gosnakego
  • Contact us

Social

  • X (Twitter)
  • GitHub
Trendshift owl
What they build speaks louder than words
© 2026 Trendshift, created by Julian Li with ❤️
SignalAdvertise
Trendshift
SignalAdvertise
Sign in
Trendshift owl

Discover trends that matter

DailyWeeklyMonthlyYearly
Live mentionsTopics
RepositoriesDevelopers
InsightsStats
Bifrost
Bifrost

shadcn/improve

Use your most capable model to audit your codebase and write plans for cheaper models to execute.

Visit GitHub
9.3k
437
5 contributors
MIT License
AchievementsView all →
FeaturedAdvertise to 180K+ monthly visitors
Bifrost

Fastest Enterprise AI Gateway for Routing, Governing, and Securing all your AI Traffic

Momentum

How fast shadcn/improve is gaining stars, and where it lands on Trendshift and GitHub Trending

Activity and trending

Stars, forks and merged pull requests each day or month, lined up with the repository's rank on Trendshift Trending

Aug 29 – Oct 28, 2026

Stars +273 in 61 days

0510
10

Forks +20 in 61 days

012
2

Merged PRs 0 in 61 days

01

Trendshift Trendingwhere the repository ranked each day it made the top 25

Not in the Trendshift top 25 in this period

Aug 29Sep 6Sep 14Sep 22Sep 30Oct 8Oct 16Oct 24
The left and right arrow keys step from day to day; Home and End go to the first and the last

Stars, forks, merged pull requests and Trendshift Trending rank, by day
DayStars gainedForks gainedMerged PRs
Sat, Aug 29, 2026+7+10
Sun, Aug 30, 2026+8+00
Mon, Aug 31, 2026+10+00
Tue, Sep 1, 2026+8+00
Wed, Sep 2, 2026+7+10
Thu, Sep 3, 2026+8+00
Fri, Sep 4, 2026+7+00
Sat, Sep 5, 2026+8+00
Sun, Sep 6, 2026+4+00
Mon, Sep 7, 2026+7+20
Tue, Sep 8, 2026+0+10
Wed, Sep 9, 2026+7+00
Thu, Sep 10, 2026+5+00
Fri, Sep 11, 2026+2+00
Sat, Sep 12, 2026+4+10
Sun, Sep 13, 2026+3+00
Mon, Sep 14, 2026+8+00
Tue, Sep 15, 2026+7+10
Wed, Sep 16, 2026+4+00
Thu, Sep 17, 2026+8+10
Fri, Sep 18, 2026+3+10
Sat, Sep 19, 2026+0+10
Sun, Sep 20, 2026+6+00
Mon, Sep 21, 2026+8+10
Tue, Sep 22, 2026+7+00
Wed, Sep 23, 2026+4+00
Thu, Sep 24, 2026+7+10
Fri, Sep 25, 2026+4+00
Sat, Sep 26, 2026+4+10
Sun, Sep 27, 2026+4+00
Mon, Sep 28, 2026+2+20
Tue, Sep 29, 2026+3+00
Wed, Sep 30, 2026+3+00
Thu, Oct 1, 2026+1+00
Fri, Oct 2, 2026+0+00
Sat, Oct 3, 2026+7+00
Sun, Oct 4, 2026+5+10
Mon, Oct 5, 2026+0+00
Tue, Oct 6, 2026+8+00
Wed, Oct 7, 2026+2+00
Thu, Oct 8, 2026+2+00
Fri, Oct 9, 2026+4+00
Sat, Oct 10, 2026+1+00
Sun, Oct 11, 2026+0+00
Mon, Oct 12, 2026+5+00
Tue, Oct 13, 2026+5+00
Wed, Oct 14, 2026+2+00
Thu, Oct 15, 2026+3+00
Fri, Oct 16, 2026+0+00
Sat, Oct 17, 2026+1+00
Sun, Oct 18, 2026+1+00
Mon, Oct 19, 2026+3+10
Tue, Oct 20, 2026+0+10
Wed, Oct 21, 2026+5+10
Thu, Oct 22, 2026+5+00
Fri, Oct 23, 2026+5+00
Sat, Oct 24, 2026+8+00
Sun, Oct 25, 2026+5+00
Mon, Oct 26, 2026+9+00
Tue, Oct 27, 2026+6+10
Wed, Oct 28, 2026+3+00
Total+273+200
  • 1#1 on the list
  • 7Rank that day
  • Not trending

    • Mentions and achievements

    Where people are talking about shadcn/improve, and every badge it has earned
    1. Sep 241 post

      • 道用最强模型干最烧脑的活、再让便宜模型去落地执行，这条路子不香吗？Shadcn 搞的 improve 技能就是这么个思路，正好戳中我之前琢磨过的方向： 🔗 说穿了就一句话：把智力最值钱的部分——啃代码库、判断值不值得做、写方案——交给最能打的模型，执行环节则丢给更省钱的模型。skill 自己一行源码都不动，计划才是它交付的东西。 主 Skill…

        Read full on X, posted by @Manorgw

        @Manorgw · x.com

    2. 9 days earlier
    3. Sep 151 post

      • • Plans are self-contained with code excerpts, verification gates, and STOP conditions • Runs focused audits for security, perf, tests, bugs, or branch changes • Executes plans via cheaper subagents…

        Read full on X, posted by @the_osps

        @the_osps · x.com

    4. 12 days earlier
    5. Sep 31 post

      • • Audits your codebase with a powerful model • Generates execution plans for cheaper models • Optimizes cost by delegating tasks to less expensive models

        Read full on X, posted by @the_osps

        @the_osps · x.com

    6. 12 days earlier
    7. Aug 221 post

      • Shadcn作者搞出来的这个improve技能，设计思路是不是让人眼前一亮？反正我之前就琢磨过类似的玩法： 🔗 它的核心逻辑是把最烧脑的活甩给最强模型,比如啃代码库、定位问题、出高质量计划，然后把动手执行那部分扔给便宜模型去跑。skill本身绝对不动代码，只负责出计划。 主Skill内置了多种运行模式，有轻量快跑的，也有全方位扫查的，这套思路不光能用在代码审计上，开发新功能也能直接套。

        Read full on X, posted by @bkdgiffug

        @bkdgiffug · x.com

    8. 14 days earlier
    9. Aug 81 post

      • Current status: running /improve and /ultrareview in Claude on everything I can.

        Read full on X, posted by @twid

        @twid · x.com

    10. Aug 71 post

      • 发现Shadcn这个improve项目，我一看到就拍大腿——跟我之前脑子里转过的一个念头简直像得离谱。 🔗 它这套打法的精髓在于：让最顶级的模型干最费脑子的活儿，比如吃透整个代码库、判断哪些地方值得动、写出详尽的实施方案；而真正动手改代码这种体力输出，全扔给便宜模型去跑。 skill自己严格遵守一条铁律：绝不动源码半行，只产出计划文档。…

        Read full on X, posted by @Gas1688

        @Gas1688 · x.com

    11. 9 days earlier
    12. Jul 291 post

      • Shadcn搞的这个improve技能，思路确实挺戳我，我之前就琢磨过差不多的玩法： 🔗 核心就一句：拿最能打的模型去扛最烧脑的活——比如扒代码库、揪问题、写高质量方案，然后把敲代码的体力活扔给便宜模型去干。skill自己绝不碰源码，只管吐计划。 主Skill配了好几种模式，有快跑也有全覆盖的，这套路子不止能用在代码审计上，搞新功能一样能套。

        Read full on X, posted by @bkdgiffug

        @bkdgiffug · x.com

    13. 13 days earlier
    14. Jul 161 post

      • First thing after upgrading is to use @shadcn /improve skill with Fable 5. It's just mind-blowing and very helpful.

        Read full on X, posted by @rnkdsh

        @rnkdsh · x.com

    15. 20 days earlier
    16. Jun 261 post

      • shadcn quietly shipped a skill that uses your smartest model to audit code, then hands the cheap model a plan to ship it. It never writes the code itself. The plan is the product. → Stamps every plan…

        Read full on X, posted by @NiteshTechAI

        @NiteshTechAI · x.com

    17. 7 days earlier
    18. Jun 191 post

      • • Audits across nine categories with file:line evidence • Plans are self-contained markdown with verification gates • Execute dispatches a cheaper model in an isolated worktree

        Read full on X, posted by @the_osps

        @the_osps · x.com

    You might also like

    Repositories close to shadcn/improve
    FeaturedAdvertise to 180K+ monthly visitors
    Bifrost

    Fastest Enterprise AI Gateway for Routing, Governing, and Securing all your AI Traffic

    anightmonarch/codex-bridge-chatgpt

    JavaScript

    Verified Codex-to-ChatGPT Web reasoning handoffs with local edits, tests, and receipts.

    14932

    adamjgmiller/adamsreview

    Shell

    Multi-lens code review pipeline for Claude Code: deep review (Claude or Codex), auto-fix loop, interactive walkthrough, external-finding injection.

    2457

    cathrynlavery/codex-skill

    Python

    Give Claude Code a second opinion using OpenAI Codex - automatic plan review via hooks

    21923

    yohey-w/codd-dev

    Python

    CoDD: Coherence-Driven Development — Claude Code plugin for cross-artifact change impact analysis

    1159

    tshemsedinov/reslop

    JavaScript

    👁️ Review AI generated code and plan changes

    1889

    mhattingpete/claude-skills-marketplace

    HTML

    Claude Code Skills for software engineering workflows - Git automation, testing, and code review

    689109

    promptadvisers/claudex

    Shell

    Autonomous Claude + Codex review loop for Claude Code. Plan mode + read-only review mode, all in one window.

    16350

    alexzh3/codex-orchestrator

    Python

    Claude Code plugin that orchestrates GPT Codex agents for implementation, monitoring, and independent reviews.

    919

    thtskaran/claude-skills

    Python

    Battle-tested Claude Code skills for hardening vibe-coded projects. Audit, clean, and ship AI-generated codebases with confidence.

    222