wangminle

Sub-agent skill for monitoring batch transcription progress and sending periodic notifications via message tool. Binds to a task_group_id, polls status at fixed intervals, and sends progress/completion/error notifications using fixed templates. Designed to run as a delegated sub-agent so the main agent can continue accepting new tasks.

Recognize transcription intent and orchestrate task submission when audio/video files or ASR keywords appear in a channel. Use when: audio/video files arrive, user mentions transcription/ASR/subtitle, user asks to batch-process media, or agent needs to proactively guide a user into the transcription pipeline.

Emergency stop workflow for FunASR Task Manager. Use when the user asks to immediately stop running transcription tasks, release occupied server slots, clear zombie TRANSCRIBING/DISPATCHED segments, diagnose active slot usage, or recover from scheduler pollution after canceled batches.

Bootstrap the funasr-task-manager development environment from scratch. Use when: first-time setup, checking if repo is cloned, installing dependencies (Python/Unicorn or Docker), or starting backend/frontend services for the first time.

Scan server-local directories for audio/video files, batch-submit them for transcription, monitor progress with proactive feedback, archive results, and handle retries. Use when: user requests batch transcription of local files, mentions scanning directories, inbox, or wants to retry failed items.

Pre-check audio/video files before transcription submission. Use when: validating whether a file can be transcribed, diagnosing file metadata or duration issues, estimating processing time before committing, or checking ffmpeg/ffprobe availability.

Reset the funasr-task-manager local database and storage to a clean state. Use when: preparing for tests (pytest, E2E, benchmark), user asks to clear tasks/data/uploads, wipe the database, rebuild after corruption, free disk space, or reset to initial state. Trigger on: "清空数据库", "重置数据", "清空任务", "只保留服务器", "其他都清空", "彻底清理", "释放空间", "reset db", "clean state", "wipe data".

Monitor transcription tasks and deliver results with quality checks. Use when: task_id or task_group_id needs monitoring or result return, user asks to export txt/json/srt/zip, user requests quality check on transcription output, or results need re-delivery for an existing task.

家庭藏书入库技能。当用户发送书封照片、ISBN、或说「买了本书/入库/加书」时使用。调用 bookshelf CLI 完成识别、元数据补全与落库。

家庭藏书查询技能。当用户问「有没有 XX 书」「我家关于历史的书」「这本书详情」时使用。调用 bookshelf find / show 检索书架。

家庭书架初始化技能。当用户说「第一次使用」「初始化书架」「怎么配置」「连接不上」「setup」时使用。引导完成 API 连通、Google Books Key、Agent 对接与成员渠道绑定。

读书笔记技能。当用户说「记一段摘录」「写点感想」「给这本书做个笔记」时使用。调用 bookshelf note 写入 reading_notes。

购书记录技能。当用户说「这本花了 X 元」「在京东买的」「补录购买信息」时使用。调用 bookshelf purchase 写入购买记录。

阅读进度跟踪技能。当用户说「读到第 X 页」「在读」「读完了」「给这本书打 5 分」时使用。调用 bookshelf progress 更新进度。

藏书统计技能。当用户问「我家有多少书」「今年买书花了多少」「阅读 streak」时使用。调用 bookshelf stats。

Analyze individual PRD documents across 6 dimensions with vision support for images. Use when (1) analyzing a PRD document's core problem, boundaries, and boundary-external issues, (2) tracking whether issues are resolved in later versions, (3) extracting key parameters and solution highlights from technical PRDs, (4) scoring document quality, (5) understanding flowcharts, architecture diagrams, UI mockups, or screenshots embedded in PRD documents. Also use when a user mentions 'per-doc analysis', 'boundary analysis', 'issue tracking', 'document review', or 'PRD quality scoring'.

Classify and overview a batch of Markdown PRD documents by category, build version evolution chains, and detect inter-document dependencies. Use when categorizing PRD or requirement documents by topic, extracting version numbers, finding document relationships, or generating overview summaries. Also use when you need to understand the structure of a document set before deep analysis, or when a user mentions 'overview', 'categorize', 'classify documents', 'version chain', or 'document relationships'.

Convert DOCX to Markdown with embedded Excel table conversion and image extraction. Use when (1) converting .docx files to Markdown, especially those containing embedded Excel spreadsheets, (2) extracting images from Word documents to local files with relative paths, (3) batch processing multiple DOCX files, (4) optionally converting Markdown to PDF with Chinese font support.

Generate structured Markdown and PDF reports from analysis results. Use when (1) rendering per-document analysis results into a readable report, (2) generating a full system review report from 7-dimension analysis, (3) producing a next-directions report from evolution tracking and gap analysis, (4) generating a PM development suggestion report with scoring cards, (5) producing a PRD draft report based on historical analysis, (6) creating Mermaid diagrams for evolution chains and dependency graphs, (7) converting Markdown reports to PDF. Also use when a user mentions 'generate report', 'export report', 'PDF report', 'review report', or 'analysis report'.

Discover requirement insights through evolution tracking and gap analysis. Use when (1) tracking how boundary-external issues are resolved across version chains, (2) building feature coverage matrices to identify uncovered functionality, (3) visualizing requirement evolution with Mermaid diagrams, (4) finding requirement gaps and overlaps, (5) generating next-direction recommendations based on unresolved issues and missing features. Also use when a user mentions 'evolution tracking', 'gap analysis', 'coverage matrix', 'requirement insights', 'unresolved issues', or 'feature coverage'.

Systematic 7-dimension review of a batch of PRD documents with multiple output modes. Use when (1) conducting a full system-level review across business value, architecture, competition, product strategy, tech evolution, PM assessment, and action planning, (2) generating next-direction recommendations for a requirement set, (3) evaluating PM writing quality and product thinking from requirement documents, (4) drafting a new PRD based on historical analysis of related documents, (5) producing any of these outputs: full_report, next_directions, quality_assessment, prd_draft. Also use when a user mentions 'system review', '7-dimension review', 'PM assessment', 'requirement review report', 'quality evaluation', or 'PRD draft generation'.

Use when a user wants to create, standardize, migrate, validate, or maintain a project-level task-list.md for software projects, agent projects, skills, workflows, planning docs, or AI collaboration rules.

Convert PDFs, especially research papers and technical reports, into structured Markdown with extracted figure/table screenshots, or prepare figure-aware reading summaries. Use for "PDF 转 Markdown", "转 md", "提取图表", "论文摘要", "带图摘要", "阅读笔记", "论文阅读摘要", "资料库入库", "处理这篇论文", "extract figures/tables from PDF", and "paper summary with figures". Do not use for generic PDF merge/split/rotate/watermark/encrypt/form-filling tasks; use the system pdf skill for those.

Use when saving web articles, WeChat posts, public Notion pages, Wiki/Docs sites, or JS-protected pages as Markdown with local images; batch crawling and merging URL lists; downloading webpage images; or converting HTML tables, code blocks, and rich text into clean Markdown.

FunASR speech recognition via WebSocket. Convert audio/video files to text using a FunASR server. Use when: (1) user needs speech-to-text or audio transcription, (2) connecting to a FunASR service for ASR, (3) batch transcribing audio/video files, (4) testing FunASR server connectivity and capabilities, (5) generating subtitles from audio, (6) measuring FunASR server performance/speed, (7) inspecting audio/video file metadata (duration, format, codec). Requires a running FunASR WebSocket server (local Docker or public test service at www.funasr.com:10096) and Python websockets>=10.0.

Convert DOCX to Markdown with embedded Excel table conversion and image extraction. Use when (1) converting .docx files to Markdown, especially those containing embedded Excel spreadsheets, (2) extracting images from Word documents to local files with relative paths, (3) batch processing multiple DOCX files, (4) optionally converting Markdown to PDF with Chinese font support.