cxin21
UserLocal-first ComfyUI mega-skill for Claude Code — 80 model prompt recipes, 662 workflow templates, hardware-aware model selection, self-updating knowledge substrate, manga end-to-end pipeline.
Categories
Indexed Skills (13)
ffmpeg-pipeline
AI 漫剧 Stage 5 — 字幕 + 视频拼接。Bash + ffmpeg CLI，零 MCP 依赖。从 Stage 4 mp4 + 对话文本生成 final.mp4。Also load chenxin-core first for VRAM/recipe context.
lora-trainer
Anima LoRA 训练编排 (v2.2) — 单路径：gazingstars123/Anima-Standalone-Trainer 独立 venv。8GB VRAM 友好，���需 ComfyUI 在线。Also load chenxin-core first for VRAM/recipe context.
manga-orchestrator
AI 漫剧自驱动 orchestrator — 6 阶段流水线自动串联。从用户触发（'全自动生成漫剧 [书名]'）到最终视频，全程零人工干预。Also load chenxin-core first for VRAM/recipe context.
manga-stage-2-panels
AI 漫剧 Stage 2 — 分镜面板自动生成。锁定 AnimaStandardV7 工作流，仅修改 prompt + LoRA 配置，自动生成 PNG 面板。可选 IP-Adapter / ControlNet 增强（v2.0）。Also load chenxin-core first for VRAM/recipe context.
manga-stage-3-review
AI 漫剧 Stage 3 — 像素级审查。对 Stage 2 的 panels 调 aesthetic-judge 逐张评分（6 维），生成双层 review.md 报告，< 7.0 自动 re-do 1 次，标 failed + redo_list。Also load chenxin-core first for VRAM/recipe context.
manga-stage-4-motion
AI 漫剧 Stage 4 — 视频生成（含说话场景）。主路锁定 ltx23AllInOneWorkflowForRTX_v44 工作流，统一处理所有视频：微动作、说话、环境音。备选 I2V_InfiniteTalk_Wan21（唇型驱动）。Also load chenxin-core first for VRAM/recipe context.
chenxin-core
Use this skill whenever a ComfyUI / generative-model request comes in for Claude Code. Trigger on ANY of: "comfyui", "comfy ui", "workflow", "出图", "跑工作流", "生成图片", "生成视频", "出视频", "manga", "漫画", "anime", "anima", "wan", "ltx", "ltx-2.3", "hunyuan", "flux", "sdxl", "sd 1.5", "sd1.5", "stable diffusion", "krea", "seedream", "nano banana", "Qwen-Image", "ideogram", "Recraft", "Kling", "Seedance", "Veo", "Sora", "Runway", "Luma", "Stable Audio", "ACE-Step", "video", "talking head", "inpaint", "upscale", "controlnet", "IP-Adapter", "refiner", "LoRA", "8 GB VRAM", "8GB", "small VRAM", "low VRAM", "vae", "unload model". This is the L4 mega-skill — it owns the dispatch table from a single Claude Code entry point to L1 (ComfyUI runtime), L2 (MCP driver), L3 (knowledge substrate: 80 recipes + 662 templates + 8 GB hardware matrix), and L5 (application skills, future P1.1). For ANY generative-model prompt, read this skill first to find the right tool, recipe, and VRAM-safe defaults before writing the prompt or invoking a w
camera-multiview
Use proactively when the user asks for a fixed character multiview set (multi-pose character sheet, three-view, reference-sheet) from a full-body image and a face image. Runs the locked Flux2-Klein workflow through local ComfyUI MCP - load this skill immediately on any such request rather than improvising with generic image-generation tools.
prompt-forge
INVOKE THIS SKILL BEFORE hand-crafting any image or video prompt. Even a one-line user brief ("写个 Anima 的旗袍提示词") MUST go through compile(spec, dialect_id); you may summarize or trim the returned PromptPackage for the user, but the validator MUST run. Skip ONLY when the user explicitly says "no skill, raw text only" / "no validation" / "just write the prompt, don't run anything". LLM-first prompt authoring and quality-audit for image and video models. The LLM authors typed concept objects (Subject with identity / appearance / pose / gesture / expression / gaze / micro_action / costume / props; Costume with garment / material / color / condition / fit / details; Prop with item / material / condition / details; Environment with place / spatial / immediate surroundings / ambient / atmosphere; Atmosphere with haze / particles_foreground / particles_midground / particles_background / wind / sky; Lighting with key / fill / rim / practical / quality / shadow_density / contrast; Frame with shot / camera_height / camer
anima-prompt-v1
Create explainable, copyable prompts for Anima-Base, Anima-Aesthetic, and Anima-Turbo with typed intent, mandatory variant quality policy, provenance-aware Catalog retrieval, explicit route selection, independent positive and negative authoring, immutable drafts, read-only inspection, and post-authoring relation submissions. Use for Anima prompt creation, refinement, tag lookup, and relation-aware Catalog maintenance.
camera-image
Execute the fixed ComfyUI Anima still-image workflow for text-to-image and image-to-image runs.
minimax-h3-prompt
Write and audit model-native MiniMax-H3 text-to-video-with-audio and reference-to-video-with-audio prompts. Use for shot timing, actions, camera transitions, reference ownership, dialogue, sound, and music; never for ComfyUI execution.
camera-video
Execute fixed MiniMax H3 text-to-video and reference-to-video ComfyUI workflows.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.