ai-shifu
- 22 skills
- 0 followers
- 2 days ago last updated
- ▌ Markdownflow Controlled Sync · ai-shifuUse when a controlled MarkdownFlowEditor integration in ai-shifu retains intermediate variable names, temporary tags, or echoed local input.
- ▌ Module Augmentation Guardrails · ai-shifuUse when TypeScript subpath exports lose types or have declaration conflicts. Follow module augmentation conventions to avoid replacing upstream exports.
- ▌ Interaction User Input Defaults · ai-shifuUse when synchronizing interactive Markdown state between ai-shifu and markdown-flow-ui. Persist raw user_input, pass userInput to ContentRender or MarkdownFlow, and let markdown-flow-ui derive button, input, and selected-option defaults instead of parsing them again in business code.
- ▌ Listen Mode Slide Mobile Integration · ai-shifuUse when integrating new markdown-flow-ui mobile player capabilities into learner listen mode. Cover orientation state, custom landscape headers, and localized player text.
- ▌ Chat System Interaction Button Overrides · ai-shifuUse when system interaction buttons must remain repeatable despite historical input or renderer readonly state.
- ▌ User Auth Flows · ai-shifuUse when changing backend user auth, verification codes, token persistence, temp-user behavior, or auth-provider integration. Keep provider dispatch and credential state centralized.
- ▌ Admin Filter Layout · ai-shifuUse when changing admin filter layouts or consolidating filter controls into the shared AdminFilter component.
- ▌ Shared Loading Dots · ai-shifuUse when adding shared Cook Web loading animations. Follow shared component placement, nonbreaking exports, and configurable dot animation conventions.
- ▌ Shifu Authoring Flow · ai-shifuUse when changing backend shifu authoring, draft history, publish, outline structure, or import/export behavior. Keep draft and publish flows consistent and protect outline integrity.
- ▌ Chat Element Streaming · ai-shifuUse when evolving ai-shifu chat streams from blocks to elements or aligning historical messages with SSE rendering. Use element_bid as the rendering key, support legacy fields, and consolidate AskBlock merging.
- ▌ Next Build Node Runtime · ai-shifuUse when Cook Web npm run build fails with SyntaxError Unexpected token '?' or next/dist/compiled errors. Check for mismatched Node runtime versions first.
- ▌ Fullscreen Dialog Portal · ai-shifuUse when Cook Web payment dialogs, settings dialogs, or other overlays appear during browser fullscreen. Check whether their portal container is outside the fullscreen element.
- ▌ Admin Table Visual System · ai-shifuUse when changing admin table presentation, sticky actions, pagination, or shared table components.
- ▌ App Error Boundary Display · ai-shifuUse when changing Next.js App Router global or route error fallbacks in Cook Web. Show the error name, message, digest, cause, URL, and available stack directly instead of only referring users to the console.
- ▌ Deep Link Lessonid Routing · ai-shifuUse when routing Cook Web deep links to lessons consistently across learner and admin surfaces. Use lessonid, reuse table-of-contents click handling, cover login, payment, and invalid-lesson fallbacks, and diagnose debug=1 initialization and authentication recovery.
- ▌ Chat Layout Width Detection · ai-shifuUse when ai-shifu chat layouts disagree between mobile and desktop. Measure the visible viewport rather than only the root element's clientWidth, update frameLayout on resize and visualViewport.resize, and clear stale mobile drawers after breakpoint changes.
- ▌ Listen Mode Audio Streaming · ai-shifuUse for learning-mode initialization, URL and stored preferences, listen-mode audio streaming, buffering, TTS request gates, and continuous playback.
- ▌ Async Confirm Dialog Loading · ai-shifuUse for Cook Web confirmation dialogs that trigger asynchronous restart, deletion, or restore requests. Prevent duplicate clicks, show loading state, and close the dialog only when the request completes.
- ▌ Chat Actionbar Ask Placement · ai-shifuUse when changing chat action bars, follow-up entry points, or AskBlock anchors. Keep content and actions synchronized and prevent duplicate inputs, empty menus, or misplaced controls.
- ▌ AI Shifu Course Creator · ai-shifu bundleUse when the user works with AI-Shifu (AI师傅) courses in any capacity of creating, writing, editing, rewriting, optimizing, reordering, deploying, publishing, previewing, or managing Teaching Prompts (per-lesson) and Course Prompts (course-level) — both written in MarkdownFlow (MDF). Covers the full course lifecycle — from converting raw material into structured lessons, to authoring interactions (single-select, multi-select, input, branching), adding variables, images, and course prompts, to deploying and managing live courses on the AI-Shifu platform. Also covers post-deployment analytics on those courses — learner count, completion rate, stuck lessons, orders, revenue, ratings, credit consumption, audience profiles, and individual learner tracking. Trigger on any mention of AI-Shifu, AI师傅, MarkdownFlow, Teaching Prompt, Course Prompt authoring, course analytics, creator analytics, 学习人数, 完成率, 卡课节, 订单收入, 积分消耗, or learner progress.
- ▌ AI Shifu Learning Report · ai-shifu bundleCreate a polished, printable learning report for one AI-Shifu course from live course analytics or a supplied report dataset. Use this skill whenever a teacher or teaching manager asks for an AI-Shifu learning report, course review, teaching diagnosis, lesson health analysis, learner engagement or audience insights, follow-up question themes, or a management-ready course analytics dashboard—even when they only say “复盘这门课” or “做个教学报告.” Produce privacy-safe `course-learning-report.json` and `course-learning-report.html`; do not use this skill for multi-course comparison or course-authoring changes.
- ▌ Course Direction Advisor · ai-shifu bundleTurn user-provided source materials into market-fit course-topic decisions without exceeding the evidence boundary of the materials. Use when the user needs course topic selection, competitor analysis, pricing guidance, audience targeting, market positioning, or a decision on whether the material is strong and complete enough to support a course. Avoid when the topic is already fixed and the user only needs lesson breakdowns, scripts, or copy polishing.