CY-CHENYUE/martial-arts-director-cy
武术指导 Skill — 一句话需求生成 N 宫格武打分镜海报与视频提示词,覆盖武术、格斗、修真、戏曲、机甲等多类题材
武术指导 Skill — 一句话需求生成 N 宫格武打分镜海报与视频提示词,覆盖武术、格斗、修真、戏曲、机甲等多类题材
Agent-first ESP-IDF setup, build, flash, post-flash verification, and troubleshooting for macOS and Windows
EasyInput V2.0 开发板的硬件资料、板级安全边界与 AI 开发上下文
EasyInput V2.0 完整固件基线,面向 AI 修改与可验证共创
File-based multi-agent project workflow for Claude Code and Codex
微信公众号文章多模板排版技能
ComfyUI nodes for Janus-Pro, a unified multimodal understanding and generation framework.
视频复刻生产 SOP Skill,基于参考视频按固定 7 步流程复刻出完整分镜包
ComfyUI custom nodes for MiniCPM
小红书自动化工具 - Playwright 驱动的小红书平台自动化操作
面向 Vibe coding / AI 编码的工程根因诊断 Skill:停止补丁循环,用系统模型、边界取证、可证伪假设与因果链定位真正故障来源(不是 PR/diff review)。
深度调研 skill:多轮搜索、交叉验证、观点聚合分析
调研报告事实核查 skill:URL 验证 + CRAAP + 四维三角验证 + AI 幻觉 7 项检查
ComfyUI InpaintEasy is a set of optimized local repainting (Inpaint) nodes that provide a simpler and more powerful local repainting workflow. It makes local repainting work easier and more efficient with intelligent cropping and merging functions.
中文视频分镜资产 prompt 工作流:参考资产 → 15 宫格(5×3)导演分镜板 → 按 P01–P15 执行的视频生成提示词
OpenClaw 技能:通过飞书视频会议呼叫用户,用于紧急提醒通知
自适应 Git 工作流 skill:支持单人直接 main、短分支、团队 PR 和已有 Git Flow
闲鱼自动回复助手 - Claude Code / OpenClaw Skill
电影分镜拓展提示词 - Skill:根据参考图片扩展出3个不同叙事功能的分镜提示词
A ComfyUI custom node that provides fine-grained control over style transfer using Redux style models.
将文本内容(字幕/文稿/大纲)转换为专业剧本和分镜脚本 Skill
Added MiniCPM language model to ComfyUI to realize advanced text generation and image understanding functions.
AIGC 营销内容生产流水线 · Claude Code skill
B站数据获取 Skill — 视频/字幕/弹幕/评论/搜索/用户/热门
Generate detailed image descriptions and analysis using Molmo models in ComfyUI.
browsermcp.io 浏览器 MCP 统一入口 skill — 一站式安装/配置/诊断/使用,复用真实 Chrome 登录态,覆盖 macOS·Windows × Claude Desktop·Code·Codex CLI
OpenClaw 图片生成技能 - 基于字节跳动 Seedream API
提示词扩写器 skill — 把模糊需求加工成可直接使用的 spec(Anthropic 三段式),≤3 问澄清,范例驱动,模型无关