Create a delightful, unexpected "wow" experience for the user by dynamically discovering and creatively combining other enabled skills. Triggers when the user says "surprise me" or any request expressing a desire for an unexpected creative showcase. Also triggers when the user is bored, wants inspiration, or asks for "something interesting".
This skill creates a delightful, unexpected "wow" experience for the user by dynamically discovering the other enabled skills in the environment and creatively combining one to three of them into a single cohesive deliverable. It solves the open-ended problem of "surprise me" or "show me something interesting" requests, where the user wants an inspiring creative showcase rather than a specific task, by turning the available skill set into raw material for a curated mini creative project.
Its workflow is a simple five-step process: discover the available skills, plan a creative mashup that juxtaposes skills in unexpected ways (such as a presentation about algorithmic art or a research report turned into a slide deck), read the full instructions of each selected skill, execute and combine their outputs into one polished artifact, and reveal it with minimal spoilers so the surprise lands. It emphasizes visual impact and emotional delight over information density, and can weave in the user's known context. When no other skills are available, it falls back to a news-based surprise, an interactive single-page HTML experience (generative art, a mini-game, a visual poem, an animated infographic, or an interactive story), or a personalized artifact.
The target users are anyone who wants inspiration, a break from routine, or a demonstration of what the assistant and its skill ecosystem can produce. It is best suited to exploratory, playful moments rather than precise deliverables, and its output quality depends entirely on which other skills happen to be enabled.
When the user says "surprise me" or expresses a desire for an unexpected creative showcase, and also when the user is bored, wants inspiration, or asks for "something interesting".
It reads the list of available skills, selects one to three of them, and designs a creative mashup aimed at a single cohesive, visually striking deliverable rather than separate demos.
It uses a fallback: a news-based designed HTML artifact, an interactive single-page web experience such as generative art or a mini-game, or a personalized artifact built from known user context.
No. By design it reveals the result with minimal spoilers — a short teaser line followed by the artifact — to preserve the surprise.
Largely the other skills available in the environment, since it combines their capabilities. With a rich skill set it can juxtapose several; with none, it relies on its built-in creative fallbacks.
Quick Setup:
.claude/skills/Repository
bytedance/deer-flow