Design Loop — Design skill for Claude Code
Run the full design iteration loop (harness component 06): planner → generator → four machine gates → critic, up to --max-rounds (default 5).
How to install Design Loop
Installs to ~/.claude/commands/tbusos-sky-skills-design-loop.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/TbusOS/sky-skills/HEAD/.claude/commands/design-loop.md -o ~/.claude/commands/tbusos-sky-skills-design-loop.md Restart Claude Code, or start a new session, for it to be picked up.
What Design Loop does
description: "Run the full design iteration loop (harness component 06): planner → generator → four machine gates → critic, up to --max-rounds (default 5). Verdict ≥ 88 ships; anything less feeds the critic's issues into the next round; rounds exhausted escalates to a human with the score trajectory. Accepts a brief + --skill (required), optional --page / --max-rounds." argument-hint: " --skill= [--page=] [--max-rou
Alternatives in Design
- Od Contribute — Open a first-contribution PR (or bug issue) on nexu-io/open-design — works for non-coders too 91.8k ★
- Design Plugin — A Claude Code plugin that helps you make confident UI design decisions through rapid iteration 674 ★
- Hn Rewrite — Read {{arg1}} (either a file path or GitHub issue URL) and rewrite it in a technical, matter-of-fact style sui 570 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
DecoupleFlow
DecoupleFlow brings lightweight soft orchestration to your existing agents. Its independent visual state-machi
Tldr Commit
Generate tldr-style commit message (verdict first, ≤50 char subject, why over what)
Utility
CSS utility class creation and updates. Use when adding, modifying, or checking Dialtone CSS utilities (d- cla
Branding Harness
Claude Code skill: generate an integrated brand identity system (strategy + verbal + visual direction) with a
Diverge
Run parallel divergent ideation on an open-ended design problem before /plan. Spawns 6 frame generators + crit
Design Critic
Use when you want an honest read on a design instead of encouragement. What is wrong, ranked by severity, with
Related Agents
Debate Judge
L0.97 Adversarial Verification — reads all 3 rounds of prosecution and defense transcripts plus the original d
Data Pipeline
Designs and builds data pipelines, ETL patterns, and API integrations. Use for anything involving fetching, tr
Codex Cross Reviewer
Cross-review orchestrator using Codex CLI as 2nd reviewer. Runs structured finding exchange with consensus loo