image-video-gen
根据文字描述生成视频，一个生成图片和视频的工作流技能。依赖 skills: byted-web-search, image-generate, video-generate。注意：此 workflow 没有执行脚本，只是一个描述性的文档。
371 repo starsObserved in 1 repos
Documents
Documents skills
Browse reusable Agent Skills, each with a clear purpose and practical guidance.
video-breakdown
视频分镜拆解技能（自包含，无需外部后端）。使用 FFmpeg 预处理视频并生成分镜数据。(1) 视频 URL 直接处理 `python scripts/process_video.py "<video_url>"`；(2) 本地文件先上传 `python scripts/video_upload.py "<file_path>"` 获取 URL 后再处理。需要本机安装 FFmpeg。
371 repo starsObserved in 1 repos
Documents
video-generate
使用 video_generate.py 脚本生成视频，需要提供文件名和 prompt，可选提供首帧图片（URL或本地路径）。
371 repo starsObserved in 1 repos
Documents
datasheet-reader
Read datasheets and technical PDF documents with `pcb scan`. Use when the user gives a local PDF path or an `http(s)` datasheet/document URL, when a task requires reading, summarizing, extracting information from, or answering questions about a datasheet or technical PDF, or when a KiCad symbol / `.kicad_sym` provides a `Datasheet` property to resolve. Run `pcb scan <input>` in bash, treat stdout as the generated `.md` path, then read that markdown file.
370 repo starsObserved in 2 repos
Documents
guide-writing
Use when writing or editing guide docs — files under guide/src/
369 repo starsObserved in 1 repos
Documents
pdf-toolkit-windows
Perform PDF operations on Windows using Python. Use for merging, splitting, rotating, extracting text, and stamping PDFs. Supports batch-safe scripts with pypdf and simple CLI inputs.
367 repo starsObserved in 2 repos
Documents
pptx-auto-windows
Generate PPTX decks on Windows using bash and Python. Use when the user asks to auto-create slides, build a .pptx from a structured outline/spec, or convert HTML (optionally to PDF) before embedding into slides. Covers python-pptx workflows, template/theme reuse, and lightweight HTML-to-PDF rendering when available.
367 repo starsObserved in 2 repos
Documents
framedex
Build a portable knowledge base of your video (and eventually photo) archive across multiple SSDs or directly from an Apple Photos library. For each clip: GPS + reverse-geocoded place, speaker-diarized multi-lingual transcript with English translation, face detection + embeddings for later named-person queries, Claude/Gemma structured assessment (keep/review/cull rating + technical quality + lighting + time of day + dominant colors + audio quality + people count + keywords + notable timestamps), and prose scene description. Writes plain-text sidecars next to originals (or to a mirror tree for Photos libraries — never inside the .photoslibrary bundle) + persistent face DB. Photos-library mode also threads through albums, named persons, and keywords from the Photos database. Non-destructive, idempotent, resumable. Use whenever you want to: index videos, tag footage, organize a drive, build the video knowledge base, transcribe audio, describe clips, rate clips, find clips by location/lighting/person/keyword, generate folder summaries, identify duplicates or cull pile, index the Apple Photos library directly. Trigger phrases: 'index this drive', 'tag my videos', 'index my Photos library', 'index Apple Photos videos', 'what's on this SSD', 'rate these clips', 'find me clips of X', 'what should I cull', 'build the video knowledge base'.
367 repo starsObserved in 1 repos
Documents
tensor-diagram-videos
How to make publication-quality manim videos of tensor-diagram derivations for the Tensor Cookbook — the animation grammar, the book's notation rules, the visual style, the review workflow, and the manim pitfalls, all learned across three videos (trace_delete, kronecker_trace, softmax_jacobian) and ~40 review rounds.
367 repo starsObserved in 1 repos
Documents