harden-task-file
Harden a manifest-dev task guidance file for one-shot quality — either /define''s quality-gate/Default set or figure-out''s probe set. Iterates: orthogonality…
Design and build digital artifacts whose audience can understand, engage with and use them: interfaces, documents, presentations, charts, media and interactive experiences. Use for creating or restyling artifacts, including disposable prototypes. Combine art direction,
$ npx -y skills add doodledood/manifest-dev --skill design --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/designContext preview
The summary Claude sees to decide when to auto-load this skill.
Design and build digital artifacts whose audience can understand, engage with and use them: interfaces, documents, presentations, charts, media and interactive experiences. Use for creating or restyling artifacts, including disposable prototypes. Combine art direction,
name: design description: 'Guide audience-centered visual and interaction design through direct study of curated examples, clean hierarchy and purposeful aliveness. HTML-first across landing pages, visual explainers and slideshows. Use when creating or restyling a visual artifact, choosing a design direction, or designing a disposable prototype.' argument-hint: '<artifact, brief or existing file>' user-invocable: true
Practice audience-centered art direction, information design and interaction design. Before composing, ground a concise design brief in the audience's knowledge, needs and use context: what must they understand, feel or do, and how should color, hierarchy, density and pacing serve that? Resolve consequential gaps from available context or ask; label assumptions when proceeding without an answer. Explicit requirements and the established design system outrank this profile. Prefer HTML when the medium is open.
Aim for polished composition, strong hierarchy, low unnecessary density and selective personality. Preserve useful labels, evidence and qualifications. Linear is a cross-category craft anchor, not a mandatory skin.
**Aliveness is a strong preference, not an animation quota.** Use responsive feedback, coherent transitions and expressive motion where they serve understanding, state, atmosphere or play. Give reading and decisions room to rest; spectacle that competes with the idea has overshot.
When refining existing work, preserve its successful identity and behavior while changing the layer the feedback names.
Before composing, load the catalog for the dominant job and open relevant examples for visual study; combine catalogs only where the jobs genuinely overlap. Other artifacts keep the direction above in their appropriate medium.
| When designing | Read | |---|---| | Landing pages | `references/landing-pages.md` | | Long-form visual research or explainers | `references/long-form-explainers.md` | | Quick visual ELI5 explainers, including internal PM briefs | `references/quick-visual-explainers.md` | | HTML presentations or slideshows | `references/html-presentations.md` | | Motion, microinteractions or app-like visual treatment | `references/aliveness.md` |
Study the examples themselves, including behavior when that is the lesson; reading annotations alone is not visual study. Before composing, name what you observed and how it informs this audience's design. Borrow within the annotated scope, keeping approved, facet-only, reserve and boundary references distinct. If visual access is unavailable, state the limitation and distinguish annotation-based guidance from observation. A preferred design is not evidence of usability or comprehension, and approval does not transfer automatically to a later redesign.
Skills for agentic coding CLIs. They keep three things in your project instead of in your head: what it's becoming, what's worth doing next, and what done means here. The agent reads them, works against them, and checks the result before reporting it finished.
Harden a manifest-dev task guidance file for one-shot quality — either /define''s quality-gate/Default set or figure-out''s probe set. Iterates: orthogonality…
Analyze Claude Code sessions to learn what went right/wrong and suggest high-confidence improvements to skills. Use when asked to analyze a session, learn from…
Sync the prompt-engineering plugin from a local clone of claude-code-plugins into .claude/ so the repo is self-contained for isolated/web environments. Copies…
Regenerate the Codex distribution copy of the plugin skills and check it for drift. Run after changing anything under claude-plugins/*/skills. OpenCode and Pi…
Analyzes the current project structure and tech stack. Use when asked to explore, understand, or summarize a project. Trigger terms: project overview, analyze…
Author-side PR lifecycle babysitter and companion to review-pr. Use when the user wants to tend an existing GitHub PR through CI, review threads, description…