Back to Details

appllama-usage

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Appllama Usage Skill

Appllama 使用技能

Appllama is a library of top-grossing mobile apps — their real screens, flows, and UI patterns, with revenue and download context. The MCP makes it an agent's research floor: not just a research tool, a builder's tool. You study what already wins, then you build something better.
Pair this skill with appllama-app-design-skill for every design/implementation step — this skill tells you what to study; that one tells you how to build.
Appllama是一个收录顶级营收移动应用的资源库——包含它们的真实界面、流程和UI模式，以及营收和下载量相关背景信息。MCP使其成为Agent的研究基地：它不仅是研究工具，更是开发者的构建工具。 你可以研究已验证成功的设计，然后构建出更优秀的产品。
将本技能与appllama-app-design-skill配合使用，覆盖每个设计/实现环节——本技能告诉你需要研究什么，而那项技能告诉你如何构建。

Ground rules (read first)

基本原则（请先阅读）

  1. Start with 
    get_credits
    — it's free.     It tells you the balance, limits, and reset date. Pro includes 1,500 credits a month (they reset in full on the 1st, UTC); every other call spends 1 credit.
  2. Go deep. Design language lives in the whole journey, not a sample — walk every screen of the apps that matter for the task, images included. That is exactly what the library is for. The one thing that's against the terms is harvesting: sweeping the catalog to extract the dataset itself rather than to answer a real task. That isn't research, and it's detected server-side.
  3. Media URLs expire in ~1 hour. Download/view what you study promptly. If links died mid-task, re-request that page for fresh ones — screen ids are durable, links are not.
  4. Ignore the watermark. Every Appllama image and video carries a small Appllama watermark in the top-left corner. It is provenance, not part of the screen — don't let it skew your read of that corner (status bar, back button, title), and never reproduce it in anything you build.
  5. Pagination is sequential. Every list response carries 
    next_cursor
    ; pass it back to continue. You cannot jump to page N — and a cursor only works for the same query that minted it. If a cursor errors, drop it and restart from page one.
  6. If you hit a rate limit, wait it out. The per-minute and per-day limits sit far above real research; on the rare hit, wait the stated time — don't retry-hammer.
  7. Errors are instructions. Tool errors are written to be acted on (expired cursor → restart; out of credits → tell the user their credits reset on the 1st and they can request more in Settings → Usage).
  1. 从
    get_credits
    开始——该工具免费。     它会显示余额、使用限制和重置日期。专业版每月包含1500个积分（UTC时间每月1日全额重置）；其他每一次调用将消耗1个积分。
  2. 深入研究。 设计语言体现在完整的用户旅程中，而非单一示例——请遍历与任务相关应用的每一个界面，包括图片。这正是该资源库的用途所在。违反使用条款的行为是数据采集：遍历整个目录以提取数据集本身，而非为完成真实任务进行研究。这不属于研究行为，且会被服务器端检测到。
  3. 媒体URL约1小时后过期。 请及时下载/查看你研究的内容。如果任务进行中链接失效，请重新请求对应页面以获取新链接——界面ID是持久有效的，但链接不是。
  4. 忽略水印。 每张Appllama的图片和视频左上角都带有一个小的Appllama水印。这是来源标识，并非界面的一部分——不要让它影响你对该区域（状态栏、返回按钮、标题）的判断，且在你构建的任何内容中都不要重现该水印。
  5. 分页为顺序式。 每个列表响应都会包含
    next_cursor
    ；将其传入下一次调用以继续查询。你无法直接跳转到第N页——且游标仅对生成它的同一查询有效。如果游标报错，请丢弃它并从第一页重新开始。
  6. 如果触发速率限制，请等待。 每分钟和每日的限制远高于真实研究的需求；如果罕见触发限制，请等待指定时长——不要反复重试。
  7. 错误即指引。 工具错误信息旨在指导你采取行动（游标过期→重新开始；积分耗尽→告知用户积分将在每月1日重置，可在设置→使用情况中申请更多）。

Tool map

工具图谱

ToolWhat it gives youTypical use
get_credits
Balance, limits, reset date. Free.Session start
search_apps
10 apps/page: name, revenue, downloads, rating, launch date, screens count, flow list with screen counts. Natural-language 
query
+ filters (revenue/downloads/rating/launch date/price/onboarding steps) + 
sort
+ 
board_id
Find the top apps for a category or need
get_app
One app in full: ratings breakdown, category rank, IAP pricing, top countries, flowsDecide if an app deserves a deep study
list_app_screens
10 screens/page in journey order (welcome → onboarding → paywall → product), each with media URL, flow, UI elements, colors. Filter by 
flow
or 
section
Walk an app screen by screen
search_screens
Screens across the whole library. 
mode="keyword"
matches screen names + filters (flow, screen_type, element, app_id); 
mode="semantic"
searches by meaning/visual language		Gather design references for one screen type
get_screen
One screen in full + up to 5 visually similar screens from other apps. Accepts 
screen_ref
= 
app_id/screen_id
(what appllama.io's "Copy Screen ID" produces)		The user pasted a screen ref; or drill into one reference
list_flows
The flow taxonomy with screen/app countsDiscover what flows exist for a category
get_flow_apps
Apps containing a flow, top revenue firstFind the best examples of one flow
list_ui_elements
~38 UI-element families with counts (one call)Vocabulary for element-level research
get_element_screens
Screens featuring an element familyStudy how winners build one component
list_my_boards
The member's own appllama.io boards (screens / apps / flows)Find their curation first
get_board
A board's full contents: screens with media, app profiles, or (app, flow) pairsWhen the member curated a board for the task, START from it
The screen_ref handshake: members can click "Copy Screen ID" on any screen at appllama.io and paste it to you. It looks like 
1393061654/spl_9i075
— feed it straight to 
get_screen(screen_ref=...)
and you're looking at exactly the screen they mean, plus its closest siblings across the library.
工具提供内容典型用途
get_credits
余额、限制、重置日期。免费。会话启动
search_apps
每页10个应用：名称、营收、下载量、评分、发布日期、界面数量、包含界面数量的流程列表。支持自然语言
query
+ 筛选条件（营收/下载量/评分/发布日期/价格/引导步骤数） + 
sort
+ 
board_id
查找特定品类或需求下的顶级应用
get_app
单个应用的完整信息：评分细分、品类排名、IAP定价、主要运营国家、流程判断某款应用是否值得深入研究
list_app_screens
每页10个界面，按用户旅程顺序排列（欢迎页→引导页→付费墙→产品页），每个界面包含媒体URL、所属流程、UI元素、颜色。可按
flow
或
section
筛选		逐界面遍历应用
search_screens
整个资源库中的界面。
mode="keyword"
匹配界面名称 + 筛选条件（流程、界面类型、元素、app_id）；
mode="semantic"
按含义/视觉语言搜索		收集特定类型界面的设计参考
get_screen
单个界面的完整信息 + 最多5个来自其他应用的视觉相似界面。接受
screen_ref
= 
app_id/screen_id
（即appllama.io上“复制界面ID”功能生成的内容）		用户粘贴了界面引用；或深入研究某一参考界面
list_flows
流程分类体系及对应界面/应用数量探索某一品类下存在哪些流程
get_flow_apps
包含指定流程的应用，按营收从高到低排序查找某一流程的最佳示例
list_ui_elements
约38种UI元素类别及对应数量（单次调用获取）为元素级研究提供术语参考
get_element_screens
包含指定元素类别的界面研究成功应用如何构建某一组件
list_my_boards
成员在appllama.io上自己创建的看板（界面/应用/流程）优先查找用户整理的内容
get_board
看板的完整内容：带媒体的界面、应用简介，或（应用、流程）配对当用户为任务整理了看板时，从该看板开始
界面引用交互： 用户可以在appllama.io上任意界面点击“复制界面ID”并粘贴给你。格式类似
1393061654/spl_9i075
——直接将其传入
get_screen(screen_ref=...)
，即可查看用户指定的界面，以及资源库中与其最相似的其他界面。

The playbooks

操作指南

ScenarioReference
Build an app from scratch (e.g. "build me a habit tracker")references/build-from-scratch.md
Make an existing screen betterreferences/improve-a-screen.md
Flow & element research; general research methodreferences/research-methods.md
Both build playbooks end the same way: the simulator loop from appllama-app-design-skill, repeated until you cannot find a flaw. Research without that loop is decoration.
场景参考文档
从零开始构建应用（例如“帮我构建一个习惯追踪应用”）references/build-from-scratch.md
优化现有界面references/improve-a-screen.md
流程与元素研究；通用研究方法references/research-methods.md
两个构建类指南的结尾方式一致：重复执行appllama-app-design-skill中的模拟器循环，直到无法发现任何缺陷。 没有该循环的研究只是表面功夫。

Local reference boards

本地参考看板

When you pull screens for study, save them into a local working structure — links expire in about an hour, but your notes and downloads don't:
research/
  <category>/
    apps.md            # the shortlist: metrics, flows, verdicts
    <app-name>/
      screens.md       # per-screen notes: id, name, flow, elements, colors
      img/             # downloaded screens, in journey order
    patterns.md        # cross-app synthesis: the category's design language
Download the screens as you study them — synthesis happens with the images side by side, not from metadata. Notes and screen IDs are durable; re-fetch a fresh link from the ID if you ever need the pixels again.
当你提取界面进行研究时，将它们保存到本地工作结构中——链接约1小时后过期，但你的笔记和下载内容不会：
research/
  <category>/
    apps.md            # 候选列表：指标、流程、评估结论
    <app-name>/
      screens.md       # 逐界面笔记：ID、名称、流程、元素、颜色
      img/             # 下载的界面，按用户旅程顺序排列
    patterns.md        # 跨应用总结：该品类的设计语言
研究时请下载界面图片——总结需要对比图片进行，而非仅依赖元数据。笔记和界面ID是持久有效的；如果后续需要图片，可通过ID重新获取新链接。