zh

研究内容为英文. 本页的研究内容（标题、简报、来源）以英文撰写，未经机器翻译。只有界面已翻译。

新想法Developer tools

Replica-powered app cloner

A hosted, no-terminal version of the exploding replica-skill pack: paste an App Store, Play or website link, and Replica's recon → architect → build → deploy skills produce a clean-room spec, a list of fixes from user reviews and a running MVP on Vercel — for founders who are not engineers.

3周前 · 评分于 2026年10月23日 · ID HBV-261023-05

暂定标题： Replica-powered app cloner (working name) — paste a hit app's link, get a clean-room spec, a "what users hate" fix list and a deployable rebranded MVP

摘要

简报假设

Benchmark: Jakeschincariol/replica-skill — total 4,846 stars (gh api, fetched 2026-10-23 07:00 BRT; browser-confirmed on github.com, logged out, read ~07:35 BRT 23 Oct: repo page shows 4.9k); created 2026-10-03 (age 20 days); 24h (22 Oct UTC) +623; 7d (16–22 Oct) +1,881 vs prior 7d (9–15 Oct) +1,733 = +9% (steady); 3d …

阅读其余部分

(20–22 Oct) +1,102 vs prior 3d (17–19 Oct) +621 = +77% (accelerating); since creation 242.3/day; peak day 2026-10-22 (+623), 0 days since peak; license MIT. Source: OSS Insight analyze-recent-stars daily (UTC) + GitHub API. Description: "Eleven free Claude skills that clone any app: reverse-engineer it, rebuild it, test it…" (recon, architect, design, build, backend, test, diff, entrepreneur, brand, launch, deploy). Forks 861. Chosen because no repo on the trending pages (trending_browser.md, captured ~06:55 BRT) is under 1 month old; it was found via GitHub search of repos created after 22 Sep 2026. Method: daily new stars from OSS Insight analyze-recent-stars (last 28 days + prior 28, UTC days) and totals from the GitHub API (gh api repos/OWNER/REPO), fetched 23 Oct 2026 ~06:59–07:20 BRT. 24h = 22 Oct UTC; 7d = 16–22 Oct vs 9–15 Oct; 3d = 20–22 Oct vs 17–19 Oct; label: >+10% accelerating, <−10% decelerating, otherwise steady; peak over the ~56 days available. Totals are GitHub API values, browser-confirmed on github.com, logged out, read ~07:35 BRT 23 Oct (github.com shows rounded totals that match); none are inferred. Existence check: Checked 23 Oct 2026: the maintainer (OpusJake, opusjake.ai) sells consulting and a free "Opus Lab", not a hosted version of replica-skill. Look-alike repos exist (Emma-Keaton/clonekit, Ayuxh-Pratap/trueclone, hello-cqq/design-clone), and hosted app/site cloners exist (same.new, Lovable, Bolt, v0, Replit Agent). No hosted product built specifically on replica-skill's 11-skill pipeline with a review-mined fix list was found. X: 91 posts/7d (Up 92% Sat→Thu (full UTC days: Sat 17 12 → Thu 22 23); peak Thu 22 23; Thu 22 is 0% below peak; window 2026-10-16 10:05–10:20 UTC (start varies by query) → 2026-10-23 10:00 UTC (X API 7-day limit; Fri 16 and Fri 23 are partial UTC days; full days Sat 17 → Thu 22)). Revenue hypothesis: ~$1K–$5K/month at month 6 (HYPOTHESIS: 50–150 runs × ~US$29).

描述假设

A hosted, no-terminal version of the exploding replica-skill pack: paste an App Store, Play or website link, and Replica's recon → architect → build → deploy skills produce a clean-room spec, a list of fixes from user reviews and a running MVP on Vercel — for founders who are not engineers.…

阅读其余部分

Who: Non-technical indie hackers and copycat founders (including HBV itself) who want to ship a better version of a trending app in days. Pain: replica-skill only runs inside Claude Code in a terminal; non-engineers cannot use it, and the output still needs setup and deployment. Monetization: Pay per clone run (hypothesis ~US$19–49 per app) plus a Pro plan for monthly runs; MIT license allows commercial hosting with attribution. Build: 2–3 agent-weeks (job runner for Claude Code skills in sandboxes, link intake, review scraper, Vercel deploy hook, credits); stack: Next.js on Vercel; sandboxed Claude Code / Agent SDK runners; replica-skill (MIT); app-review scraping via Apify; Stripe credits. GTM: Before/after clone showcases (faceless); indie-hacker communities; SEO for "clone [app]".

目标客户

Non-technical indie hackers and copycat founders (including HBV itself) who want to ship a better version of a trending app in days.

痛点

replica-skill only runs inside Claude Code in a terminal; non-engineers cannot use it, and the output still needs setup and deployment.

为何是这个分数

分数较低的原因： Crowded prompt-to-app market plus legal exposure for "clone any app" positioning.

护城河假设

Weak: The skills are MIT and public. Only the review-mined fix list and UX differ. Status: Hypothesis.

需求信号

X 上的帖子（7 天）

91

信号趋势

Up 92% Sat→Thu (full UTC days: Sat 17 12 → Thu 22 23); peak Thu 22 23; Thu 22 is 0% below peak

网络需求已验证

GitHub: 4,846+ stars in 20 days, accelerating (above). X: 91 posts in 7 days on cloning apps with Claude skills, Sat 17 (12) → Thu 22 (23), peak Thu 22 (Verified X API) — small.

市场与竞争对手

TAM

Not sized this run — Scope: AI app builders. Method: No report fetched. (Basis: n/a)

SAM假设

Not sized — Scope: Non-technical copycat founders. Method: HBV assumption. (Basis: Hypothesis)

SOM假设

~$12K–$60K ARR (hypothesis) — Scope: Month-6 run-rate. Method: From revenue hypothesis ×12. (Basis: Hypothesis)

竞争对手

same.new; Lovable; Bolt.new; v0; Replit Agent; clonekit, trueclone, design-clone repos; Claude Code itself.

商业模式

变现方式假设

Pay per clone run (hypothesis ~US$19–49 per app) plus a Pro plan for monthly runs; MIT license allows commercial hosting with attribution.

收入假设（每月）假设

~$1K–$5K/month at month 6 (HYPOTHESIS: 50–150 runs × ~US$29).

待完成的任务

When a new app is going viral and I cannot code, I want to get a working, better copy in a day, so I can launch before the wave passes.

HXC

Bruno, 29, marketer in Florianópolis (illustrative) · Spots trending apps on X; cannot code; has tried Lovable.

构建与 GTM

技术栈

Next.js on Vercel; sandboxed Claude Code / Agent SDK runners; replica-skill (MIT); app-review scraping via Apify; Stripe credits

GTM 渠道

Before/after clone showcases (faceless); indie-hacker communities; SEO for "clone [app]"

风险

主要风险

Crowded prompt-to-app market plus legal exposure for "clone any app" positioning.

全部风险已验证

Value: Supported — Accelerating repo (Verified API) | Usability: High — Crowded hosted builders | Feasibility: High — IP/trademark risk | Business Viability: Attention — Compute cost

下一步验证

Not recommended now. If tested: run replica-skill on one HBV copycat (e.g. a past EXISTING row) internally to measure time-to-MVP.

来源

  1. github.com
  2. opusjake.ai
  3. github.com
  4. github.com
  5. github.com
  6. ossinsight.io
  7. X API counts (window above)

ideanewgithub-wrapperapp-clonerclaude-skillsindie-hackerdeveloper-toolsglobal

想在评分当天收到下一个想法吗？

分数、来源、风险和下一步，通过邮件发送。

免费。通过 beehiiv 发送。一键退订。

  • 新想法Developer tools· 0周前

    AgentParity — continuous parity

    A CI service that replays a real agent trajectory against every self-hosted framework release and asserts interrupt/stream/checkpoint/durability parity, failing the build before a regression reaches production — self-serve, $35/repo/month.

    分数较低的原因： Adjacent platforms (LangSmith, AgentOps, promptfoo) ship parity checks for free

    打开 ProductBots.ai →

  • 新想法Developer tools· 1周前

    Embeddable 2.5D avatar player SaaS

    Hosted drop-in 2.5D lip-sync avatar widget for sites (docs, support, marketing) with API/seat billing — player infrastructure, not a creation studio.

    分数较低的原因： Owner LivLiv already monetizes hosting and a partner Web SDK; thin whitespace

    打开 ProductBots.ai →

  • 现有业务Developer tools· 1周前

    NoMac-style copycat

    NoMac.app rents disposable Apple-silicon Macs to AI agents by the minute (US$0.90/h Verified list price); TrustMRR shows about US$2,617 revenue in 30 days (+1,217%). HBV angle: a NoMac-style copycat with LATAM billing and distribution.

    分数较低的原因： Apple licence + hardware economics (fleet cost vs utilization unknown; sub-24h terms Hypothesis) and crowded category

    打开 ProductBots.ai →

这是研究，不是投资建议。数字保留其发布时的标签。