Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    微信小程序提审预检 Gate | Mini Program Review

    1

    Catch what gets a WeChat Mini Program rejected before you hit 提交审核 — undeclared privacy APIs (errno 112), 诱导分享, forced login, category mismatch, test leftovers. file:line fixes.

    $19

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    微信小程序提审预检 Gate | Mini Program Review

    微信小程序提审预检 Gate | Mini Program Review

    Example session with this skill installed

    明天提审,帮我检查一下。类目是'生活服务>综合生活服务'。隐私保护指引里只声明了位置信息。已经备案了。社区发帖功能也有。

    [project: ./my-miniprogram — 原生开发, pages: index / post / game]

    (Fictional test project.)

    • Read your context and instructions
    • Compiled the gate

    小程序提审预检报告

    服务类目: 生活服务 > 综合生活服务(开发者口述,未核实) 备案: 用户确认,未核实

    1. 结论

    BLOCK:pages/game 存在"分享到群即可复活"的奖励式诱导分享(运营规范 5.1);另有 4 项隐私接口未声明、首页强制登录且跳转到不存在的页面、内网 HTTP 接口、UGC 无内容安全、发帖和游戏超出类目。按当前代码,提审基本会被驳回,部分功能上线后也会因 errno 112 直接失效。

    PASS 表示主要驳回风险已排除,不保证审核通过。

    2. 隐私接口声明对照

    类别接口/组件位置已声明处理
    位置wx.getLocationpages/index/index.js:4是保留;requiredPrivateInfos 已含 ✔。只需城市级时改用 getFuzzyLocation
    剪切板wx.getClipboardDatapages/index/index.js:5否启动即静默读取。建议删除,或改为用户点"粘贴邀请码"后读取并声明
    相册/相机wx.chooseMediapages/post/post.js:1否声明"选中的照片或视频信息",用途:发布社区帖子
    用户信息open-type="chooseAvatar"pages/index/index.wxml:3否声明;onAvatar 未实现,需补全
    手机号open-type="getPhoneNumber"pages/index/index.wxml:2否声明;onPhone 未实现;额度耗尽时需有替代登录方式

    同意流程:官方弹窗;拒绝后未处理(无 fail 回调,门店列表无降级)。app.json 中的 __usePrivacyCheck__ 已废弃,建议删除。

    3. 运营规范问题

    级别条款位置问题修改
    BLOCK5.1 诱导分享pages/game/game.wxml:1;game.js:3文案"分享到群即可复活",且 onShareAppMessage 里直接 lives + 1解除分享与奖励的绑定:onShareAppMessage() { return { title: '一起来玩', path: '/pages/game/game' } }
    REVISE强制登录pages/index/index.js:6无手机号缓存即 redirectTo 登录页允许游客浏览,在发帖/下单时再引导登录
    REVISE功能不完善pages/index/index.js:6/pages/login/login 未注册且不存在新建并注册,或删除跳转
    REVISE功能不完善pages/post/post.js:1API 只在 index.js 中定义,发帖会报 ReferenceError放入 config.js 并在各页 require
    REVISE测试残留pages/index/index.js:1http://192.168.1.20:8080/api改为已配置的 HTTPS 合法域名
    REVISE类目 3.4/5.7pages/post;pages/game社区发帖、游戏玩法超出"综合生活服务"补充类目,或本次真正移除入口(不得用远程开关在审核期间隐藏,违反 5.8.3)
    REVISEUGC 内容安全pages/post无 msgSecCheck / mediaCheckAsync,无举报入口服务端审核后再展示;增加举报

    4. 审核备注建议

    门店列表无需登录即可浏览;发帖功能需登录,测试账号:【待填】。

    5. 限制说明

    • 服务端代码、后台配置和名称/头像未提供,未核实。
    • 规则以 2026 年 10 月为准,条款编号以官方页面为准。

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Mini Program reviews reject for the same reasons over and over: a privacy API you forgot to declare (and then fails in production with errno 112), a share-to-unlock feature, forced phone login on the home page, a feature outside your 服务类目, a "测试" string or a 192.168 URL left in the code. This skill finds them in your code before the reviewer does.

    提审前跑一遍:扫描代码里的隐私接口并和你的《用户隐私保护指引》对照,检查诱导分享、强制登录、类目不符、测试残留、UGC内容安全、备案和AI功能要求,给出具体到行的修改。

    What it checks | 检查内容

    • Privacy declarations — every privacy API/component in your code vs your 隐私保护指引, location APIs vs requiredPrivateInfos, consent flow and what happens when users decline
    • 诱导分享 (rule 5.1) — share handlers that grant rewards, unlocks or extra lives, and share-to-group wording
    • Forced login — redirects to login before users can browse
    • Category fit (3.4 / 5.7) — posting, selling, finance or games outside your approved 类目
    • Broken flows — routes to unregistered pages, missing handlers, non-HTTPS or internal-IP URLs, test leftovers
    • UGC — content-security checks (msgSecCheck / mediaCheckAsync) and a report entry
    • 备案, AI features, phone-number quota — flagged with what to confirm in the console
    • Review-mode switches — flagged, never helped (rule 5.8.3)

    What you get

    • A bundled scanner (stdlib Python) for native, uni-app and Taro projects — every hit then confirmed by reading the code
    • PASS / REVISE / BLOCK with file:line, rule number, and code fixes
    • A privacy declaration table with suggested purpose text
    • A console checklist and a draft 审核备注 with test-account instructions

    Honest limits: PASS lowers rejection risk but can't guarantee approval; reviewers use judgment. Rules current as of October 2026, with rule numbers to check against the live page.

    Can't see your WeChat admin console: 类目, 资质, 隐私保护指引 contents and 备案 are checked against what you tell it and marked 未核实. The scanner is line-by-line pattern matching; every hit is confirmed by reading the code, but server-side code (e.g., content security) is only checked if you provide it. Platform rules and API behavior as of October 2026; rule numbers should be checked against the live page. Doesn't cover mini games' separate rules in depth. PASS can't guarantee approval.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Permissions

    Terminal / Shell
    Read Files

    Runs a bundled read-only Python scanner (standard library only) over your Mini Program project and reads the source files to confirm findings. Doesn't modify files, call the network, or need any WeChat credentials or AppSecret.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    Yes. The scanner recognizes uni.xxx and Taro.xxx calls as their wx.xxx equivalents and reads uni-app's pages.json. For best results, point it at both your source and the compiled WeChat output.

    Yes. errno 112 means a privacy API isn't declared in your 用户隐私保护指引. The skill lists every privacy API in your code with file:line, compares it with what you declared, and suggests the purpose text for each.

    No. It works from your code and what you tell it about your 类目, privacy guide and 备案. Anything it can't see is marked 未核实, and you get a console checklist to confirm.

    No. Review-mode switches violate operating rule 5.8.3 and can get the account banned. If it finds one, it flags it. It helps you either add the right category or remove the feature properly.

    No. PASS means it found none of the common rejection causes in your code. Reviewers still use judgment, and rules change, so it lowers your risk rather than guaranteeing approval.

    New in Code Quality & Review