superdesigndev/superdesign-skill
Design skill for UI and marketing graphics on the Superdesign canvas: reads the repo for context, extracts its design system, then generates and iterates branchable design drafts, flow pages, and reusable components through the Superdesign CLI.
Install
via GitHub · GitHub source
Installing from GitHub runs the project's build script, which pnpm blocks until you allowlist it — run the command once and pnpm prints the exact key to add under `allowBuilds` in ~/.dsh/profiles/web/pnpm-workspace.yaml.
What happened when we ran it
Installed cleanly when we ran it
Every command here is run in a throwaway container against a clean profile, and the result is whatever the harness recorded — not a guess from the source. Last run 11d ago.
Show it in your README
For maintainers: the badge serves this listing's latest sandbox verdict, so a re-run updates it on its own — and it links readers to the full result here.
Due diligence
What it is — Superdesign-skill is a design skill that generates iterative design drafts for UI and graphics through the Superdesign CLI.
Who it is for — When iterating product feature page design in Cursor, this plugin provides support for design system extraction and draft generation. If all design tasks are handled in the superdesign.dev web app, then no need to install via plugin.
Watch out — Sandbox test passed: installation and registration into a new profile via harness succeeded. No obvious pitfalls found.
The verdict — I would install it because it supports cross-session continuity, allowing iterations to resume without repeating codebase discovery.
Generated by grok-4.6, and a starting point rather than a verdict. Where it says a plugin installs or does not, that is from a real run in a clean profile — everything else is read off the repository. Trust the source over this.
superdesign-skill is a DeepSeek Harness plugin that lets an agent design UI and marketing graphics on the Superdesign canvas, rather than generating interface code without a visual drafting stage. It reads the repository to understand the existing UI and design system, stores the system guide at .superdesign/design-system.md, and can create, iterate, branch and fetch design drafts through the superdesign CLI. For existing pages, it can also create replica templates under .superdesign/replica_html_template/<name>.html, which provide the current UI as a baseline for canvas work.
The plugin is intended for developers, product designers and indie hackers working from a coding agent who want to explore page designs, flows and reusable components before implementation. Its terminal surface reaches the repository files it needs to inspect or write and invokes the separately installed superdesign CLI; the CLI must be authenticated with superdesign login. Later sessions can resume project and draft context when the relevant codebase state has not changed.
superdesign-skill is a poor fit when a project only needs direct UI implementation, or when the Superdesign CLI and its canvas workflow are not available. The canvas does not access the codebase directly, so design work on an existing page depends on first producing an accurate replica HTML template. It is also separate from the archived Superdesign VS Code extension.
superdesign-skill is invoked through the agent and follows a design workflow rather than a single fixed command. For an existing application, it investigates the relevant UI, creates or updates .superdesign/design-system.md, gathers requirements, and asks whether to design in Superdesign or implement directly. If canvas design is selected, it creates a current-state replica at .superdesign/replica_html_template/<name>.html, then uses that replica and the design-system guide as project context. The replica should describe the current UI only; it is not the place to add the new design.
After project setup, the agent creates a design draft and iterates or branches it using the returned designDraft ID. Flow work is split into planning flow pages and executing flow pages. Existing project, draft, extracted component and source-context information can be retained across sessions.
The README documents these superdesign CLI operations:
| Command | Purpose |
|---|---|
superdesign search-prompts --query "<keyword>" |
Search the prompt library by query |
superdesign search-prompts --tags "style" |
Search by tags, optionally with --query |
superdesign get-prompts --slugs "<slug1,slug2,...>" |
Read the compact index for selected prompts |
superdesign get-prompts --slugs "<slug>" --full |
Fetch full prompt bodies |
superdesign extract-website --url https://example.com --design-md |
Extract a site's design DNA into a design guide |
superdesign login |
Authenticate the CLI |
The canvas tooling also covers project creation, optional HTML-template seeding, draft creation, draft iteration or replacement, branching, flow-page planning and execution, and fetching a specific draft. CLI output is intended to be read directly; use --json for a full machine-readable payload and --full to expand truncated fields.
The README lists support for Claude Code, Cursor, Codex and more than 70 coding agents. It requires the separately installed @superdesign/cli, followed by superdesign login. The README does not specify environment variables, operating-system constraints or CLI authentication defaults.
Written from the project's own documentation and kept in sync with it. Where the two disagree, the source is authoritative — read the README on GitHub
Same category
zhu1090093659
Skill center for the dsh web GUI: browse all loaded skills grouped by source, enable or disable model invocation, create new skills, and delete into a recoverable trash.
Installed cleanly when we ran it
chuspeeism
现代化可灵活嵌入的任务面板，支持 Codex、DeepSeek Harness
Installed cleanly when we ran it
Alisa0808
Turn one topic into a finished Vox-style paper-collage explainer/ad video — automated end to end on Atlas Cloud + ffmpeg. An agent skill.
Installed cleanly when we ran it
GanyuanRan
Software-engineering method pack for coding agents, with skills for baseline-first planning, systematic debugging, prompt hygiene, verification before completion, and repair/retirement tracking.
Installed cleanly when we ran it
jingyunstudio
基于 Jingyun Studio + DeepSeek Harness (DSH) 打造的一站式 AI 商业化桌面客户端
Did not install when we ran it
