appllama-skills

作者 Appllama已验证

A builder, not just a researcher. Agent skills that turn top-grossing app patterns into native-quality mobile screens.

97
Stars
5
Forks
2026/8/24
添加时间

⚠️ 第三方软件声明

本 Skill 为第三方开源软件,独立托管于 GitHub。SkillTip 仅为信息目录,不控制或维护底层仓库。所显示的安全检查为自动化且范围有限,安装前请自行审查源码。

阅读服务条款

安装

添加到你的 Claude Code skills 目录:

# Add to your Claude Code skills
git clone https://github.com/Appllama/appllama-skills

快速入门

使用 appllama-skills 等 Skills 的指南。

安全报告

已验证

上次扫描:—

{
  "status": "PASSED",
  "issues": []
}

README.md

Appllama is the design library of top-grossing mobile apps — their real screens, flows, and UI patterns, with revenue and download context. These skills turn that library into an agent's working method: study every screen of the apps that already win, extract the category's design language, then build screens that hold up next to them.

The skills

Skill What it does

appllama-usage The research engine: how to use the Appllama MCP like a design director — the full tool map, and the playbooks for building an app from scratch, improving an existing screen, and flow & element research.

appllama-app-design-skill The build bar: native-feeling Expo / React Native screens — Apple HIG fidelity, semantic colors, native controls, anti-slop discipline, navigation that behaves, a strict motion bar, and a full-motion simulator loop (whole flows recorded and scrubbed frame by frame, not screenshots). The reference library below has the detail.

They are designed as a pair: usage decides what to study, design decides how to build, and both insist the loop only ends in a simulator with a screen you can't fault.

Inside the design skill

A short set of laws, and a reference library the agent loads only when the task calls for it:

Reference What it settles

navigation push vs replace vs dismissTo · modal vs form sheet vs overlay · tabs and what covers the tab bar · deep links with a real stack underneath · the one-way doors (sign-in, onboarding done, purchase, finished session) · the back-stack audit

motion the decision sequence — should it animate at all → purpose → cheapest tool → properties → spring or curve → off the JS thread — with exact values, haptics, reduced motion, and the never-ship list

motion-recipes press feedback, drag-to-dismiss sheet, swipe-to-delete, collapsing header, list entrances, keyboard-synced UI, tab indicator, toast, threshold haptics — ready to build

fluid-interfaces the physics of feel: response, interruptibility, velocity hand-off, momentum projection, rubber-banding — plus materials and depth, multimodal feedback, typography, and the design principles behind all of it

motion-review reviewing a diff's motion, auditing a whole app into plans any agent can execute, and hunting for (and rejecting) places that could animate

motion-vocabulary the exact words for motion, so a brief that says "bouncy" becomes a spec that says what it means

variant-lab three genuinely different directions behind a dev-only switcher — for open briefs and hero screens where direction matters more than polish

native-controls the iOS + Android control map, menus, sheets, forms — and the library picks, so nothing solved gets hand-rolled

performance measure → fix → re-measure, the budgets, and the thread discipline behind 60 fps

image-assets one style system, generated at the highest quality, post-processed and verified in both themes

simulator-loop the verification checklist — layout, theming, motion, interaction, navigation and back stack, state — and the device matrix

Install

One command, from your project root — works with Claude Code, Cursor, Codex, and 70+ other agents:

npx skills@latest add appllama/appllama-skills

Variations:

# install for specific agents, no prompts
npx skills@latest add appllama/appllama-skills -a claude-code -a cursor -y

# install user-wide instead of per-project
npx skills@latest add appllama/appllama-skills -g

Only want the app design skill?

appllama-app-design-skill stands on its own — the native-quality build bar, anti-slop discipline, and the full-motion simulator loop work with or without the Appllama MCP connected:

npx skills@latest add appllama/appllama-skills --skill appllama-app-design-skill

(The same --skill flag installs only appllama-usage if you want just the research engine.)

Skills are plain directories — copy them into your agent's skills folder (.claude/skills/ per project, ~/.claude/skills/ user-wide, or your harness's equivalent):

git clone https://github.com/appllama/appllama-skills
cp -r appllama-skills/skills/* ~/.claude/skills/

Connect the Appllama MCP

appllama-usage runs on the Appllama MCP; appllama-app-design-skill is sharper with it connected. The endpoint:

https://mcp.appllama.io/mcp

Add it as a custom connector in Claude, Cursor, Codex, or any MCP client and approve the connection with your Appllama account. MCP access is part of Pro; credits reset in full on the 1st of each month. Every call spends one credit — get_credits is always free.

Try it

With the MCP connected and the skills installed, ask your agent:

Build me a habit tracker. Study the top-grossing habit apps first and don't stop until every screen survives the simulator comparison.

Make this screen better. (paste a screenshot, code, or a "Copy Screen ID" ref from appllama.io)

How do the best fitness apps structure onboarding — how long, what does each step earn, and where does the paywall sit?

Wire up the checkout flow. Decide which screens push, which present as sheets, and make sure nobody can go back into the paywall after paying.

Review the animations in this app — what should be deleted, what's on the wrong thread, what's missing velocity — and give me the plan.

License

MIT. The Appllama name, llama, and logo are trademarks of Antmind Ventures Private Limited — the license does not grant rights to use them.

常见问题

What is appllama-skills?

appllama-skills is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Appllama. A builder, not just a researcher. Agent skills that turn top-grossing app patterns into native-quality mobile screens. It has 97 GitHub stars.

Is appllama-skills safe to use?

Yes. appllama-skills passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.

How do I install appllama-skills?

Clone the repository with "git clone https://github.com/Appllama/appllama-skills" and add it to your Claude Code skills directory (see the Installation section above).

Are there alternatives to appllama-skills?

Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh appllama-skills against similar tools.

评论 (0)

暂无评论,成为第一个分享想法的人!

ECC

by affaan-m

10

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

242,21936,702JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情
15

An agentic skills framework & software development methodology that works.

234,96620,863Shell
AI 智能体ai-agentsbrainstorming
查看详情

hermes-agent

by NousResearch

10

The agent that grows with you

234,43747,175Python
AI 智能体ai-agentsagent-orchestration
查看详情

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

185,94028,768JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情

cc-switch

by farion1231

3

A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io

128,8688,826Rust
AI 智能体claude-codeai-tools
查看详情

claude-code

by anthropics

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

120,03119,897Shell
AI 智能体
查看详情

开发者还喜欢

基于喜欢此 Skill 的开发者投票和收藏

ECC

by affaan-m

10

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

242,21936,702JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情
15

An agentic skills framework & software development methodology that works.

234,96620,863Shell
AI 智能体ai-agentsbrainstorming
查看详情

hermes-agent

by NousResearch

10

The agent that grows with you

234,43747,175Python
AI 智能体ai-agentsagent-orchestration
查看详情

n8n

by n8n-io

12

Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.

201,88160,308TypeScript
MCP 服务器apisai-tools
查看详情

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

185,94028,768JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情

cc-switch

by farion1231

3

A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io

128,8688,826Rust
AI 智能体claude-codeai-tools
查看详情