seedance2-0:在 Agent 里调用 Seedance 2.0 做文生视频

seedance2-0 是 SkillHub 社区技能 @user_1da200f4/seedance2-0(v1.0.4,design-media),面向字节 Seedance 2.0,提供文生图/文生视频的快速生成封装。通过 skillhub install 安装到 Cursor 等 skills 目录,配置 ARK_API_KEY 与 ARK_BASE_URL 后,Agent 可按模板收集时长、比例、分辨率等参数并调用封装脚本生成与下载视频。适合已开通火山引擎 Ark 的开发者;需注意 API 密钥、内容合规及社区技能评估。

Read More
Zopia-Skill:把 AI 视频创作流程挂进 Agent

Zopia-Skill 是 SkillHub 社区技能(@user_680513bf/zopia,v1.1.0),通过 Zopia 平台 AI Agent 覆盖文生/图生视频、角色与分镜图、剧本写作及多集短剧制作。本文基于 SkillHub 官方简介说明其核心能力、触发条件(剧本→角色→分镜→视频)、SkillHub CLI 安装方式与 Cursor/Claude Code/Codex 的 skills 目录,并提醒 requires_api_key 与 --dir 配置要点。

Read More
dsh-video-creator: WeChat Channels Creation Assistant

`dsh-video-creator` is a video channel content creation assistant plugin for DeepSeek Harness, designed to use built-in templates and domestic large models to standardize the generation process of short-form video content from script planning to multi-platform publishing plans. It covers scenarios such as talking-head videos, product recommendations, and short dramas, and supports one-click generation of titles, storyboards, subtitles, and compliance-ready copy. Users need to configure model APIs such as DeepSeek. The plugin can generate publishing files, but due to platform API limitations, it currently only supports generating copy for manual copying and uploading Bilibili account credentials, and does not yet support automatic uploading to platforms such as Douyin and Kuaishou.

Read More
dsh-video-studio (Jingying): DSH Native Video/Manhua Generation Plugin

dsh-video-studio (Whale/鲸影) is a native DSH AI video and comic production plugin that integrates the comprehensive workflow of story, script, storyboard, assets, generation, and editing, supporting step-by-step execution, quota scheduling, and auditing. Its core features include a seven-stage director pipeline, where the first three stages are executed by DSH session models for LLM tasks. A multi-account quota pool offers rotation, rate limiting, backoff, and auditing functions, with credentials stored in local secure files. Four-tier prompt engineering enhances quality through local optimization and scoring feedback. Style memory and consistency asset libraries ensure unified character and scene consistency across sessions. Additionally, it supports parallel shot generation, automatic ffmpeg rendering, CapCut draft export, and optional lip-sync. The plugin provides multiple model tools and Host routing interfaces, suitable for scenarios such as mock execution with pre-baked packages, prompt optimization, video generation, and self-auditing. Please note that some providers have not verified real keys or...

Read More
dsh-xiapan-media: Adding Vision, Image Generation, and Video Generation to DeepSeek Harness

`dsh-xiapan-media` is a client plugin for DeepSeek Harness (DSH), designed to integrate visual understanding, image generation, and video generation capabilities into existing text conversations by connecting to the Xiapan cloud service. Maintained by `dongsheng123132` under the MIT license, the plugin requires Node.js >=22.19.0. Its core features include: leveraging the `xiapan-vision` route for image understanding and OCR, supporting direct image paste with backend translation into text; performing image generation and modification via the `gpt-image-2` model; and calling the Seedance service to generate 5-15 second videos. The plugin supports standalone installation via `cordis.patch.yml` or

Read More