Image · Design & Media
NeoSpark Video
Generate videos and manage NeoSpark video tasks. Text-to-video from a prompt, image-to-video from a first or last frame, multi-reference generation from reference images and videos, and watermark removal. Also lists video models, lists tasks, checks task status, collects finished videos, and deletes task records.
Try it now
Try NeoSpark Video
Pick an example or write your own. Enter or press the arrow to open it in the conversation platform, with this Skill already selected.
Start from an example
What happens when you paste it
- 1Your agent connectsIt reads the install guide and wires up the official NeoSpark CLI.
- 2It installs NeoSpark VideoThe Image package arrives with its SKILL.md, examples and references.
- 3You get the workThe Skill follows your brief and returns the image you asked for.
Works in Claude Code, Cursor, Codex and any agent that reads SKILL.md. See what it makesInstall guide
What this Skill does
Generate videos and manage NeoSpark video tasks. Text-to-video from a prompt, image-to-video from a first or last frame, multi-reference generation from reference images and videos, and watermark removal. Also lists video models, lists tasks, checks task status, collects finished videos, and deletes task records.
Social cards, illustrated notes, calendars and portraits, made to a declared format rather than a generic prompt.
Work where the art direction is the product: decks, portraits and sites with a strong visual point of view.
Install it once and the instructions, templates and scripts travel with the folder. Your agent reads SKILL.md, verifies the package, and can make this kind of work again from the next ask.
Install with one prompt
Paste this into Claude Code, Codex, Cursor or any agent that reads SKILL.md. It connects the CLI, installs NeoSpark Video and waits for your next task.
Ask your agent to install it
Read https://useneospark.com/install/agent.md and follow the instructions to connect the official CLI, then install the NeoSpark Video Skill (neospark-video-cli) and use it for my next task.Related Skills
More work in Image and Design & Media.
Image · Content & Publishing
Agent Reach
Read and search public internet content from an agent: fetch any web page as Markdown, read RSS feeds, query V2EX, Xueqiu, Bilibili, GitHub, YouTube captions, single public X/Twitter posts, and run Exa web search. Read-only. Browser-session platforms (Reddit, XiaoHongShu, LinkedIn, Boss, Facebook, Instagram, Xiaoyuzhou) and X/Twitter search are not supported.
Image · Data & Analytics
电商详情页视觉专家
为电商产品生成高端九宫格视觉策略、9 张主图提示词，并调用 NeoSpark API 生成预览图与独立主图
Image · Business Ops
视频换角色
用角色参考图替换视频中的指定人物，保留原视频的动作、镜头与背景；调用 NeoSpark wan3.0-video 异步生成并下载成片。