Reverse-engineers reference videos using FFmpeg, audio transcription, and Vision LLM analysis, then generates Seedance 2.0 prompts with optional face, body, and clothing replacement modes.
视频逆向复刻 — 分析参考视频(FFmpeg帧提取+Vision LLM) + 生成复刻Seedance 2.0 Prompt + 4种素材替换模式。支持单个和批量。Use when: '复刻这个视频', '分析爆款视频', 'replicate this video', '视频逆向', '反编译视频', '批量分析视频'. Do NOT use for creating from scratch — use fashion-video-creator instead.
Helps agents draft, edit, and audit prose for 41 patterns associated with AI-generated writing while preserving source information, matching voice, and avoiding fabricated facts.
Supports structured collaboration on multi-volume novels through commands for outlining, planning, chapter drafting, prose generation, consistency checks, and archiving tied to ex…
Automates WeChat Official Account operations, including content collection, daily and weekly analytics reports, comment monitoring and replies, and article generation workflows.
Guides fiction and nonfiction writers through planning, research, drafting, revision, creative blocks, writing habits, and publishing workflows including queries, proposals, and s…
Enables agents to inspect, transcode, filter, edit, render, and verify local audio and video with FFmpeg/FFprobe using explicit stream selection and evidence-based workflows.
Manage posts, pages, and tags on a running Ghost CMS site through its Admin API, including drafting, publishing, scheduling, editing, deletion, browsing, and site inspection.
Read, create, edit, validate, convert, and extract text, metadata, structure, images, and knowledge from EPUB2 and EPUB3 ebooks.
Guides agents in creating or analyzing Gibson-informed cyberpunk settings, scenes, world operations, and visual briefs while separating canon, inference, and original adaptation.
