explore-auto

An exploratory UI design workflow that creates multiple interface variants from prompts or images, then lets you choose targets and designs.

In plain words
What is it for?
Use it to generate variants for full pages such as dashboards or isolated components such as cards, forms, navigation bars, or hero sections.
Why use it?
It organises the repeated work of extracting styles, animations, and layouts before assembling several possible interfaces.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/catlog22/claude-code-workflow/explore-auto
Clone the repo
git clone --depth 1 https://github.com/catlog22/Claude-Code-Workflow

Made for: Claude Code.

Per session 27 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 7,384 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00027 $0.07384
Opus 5 $0.00014 $0.03692
Sonnet 5 $0.00005 $0.01477
Haiku 4.5 $0.00003 $0.00738

Measured 2d ago against content hash 19efc45fa068, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

explore-auto scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

Bash(rm -rf "{base_path}/style-extraction" "{base_path}/animation-extraction" "{base_path}/layout-extraction" 2>/dev/null)
.claude/commands/workflow/ui-design/explore-auto.md · 679 lines

How it starts

The opening of the file, as written. The whole thing — 679 lines — stays where its author put it; the contents beside it link to each section on GitHub.

UI Design Auto Workflow Command

Overview & Execution Model

Fully autonomous orchestrator: Executes all design phases sequentially from style extraction to design integration, with optional batch planning.

Unified Target System: Generates style_variants × layout_variants × targets prototypes, where targets can be:

  • Pages (full-page layouts): home, dashboard, settings, etc.
  • Components (isolated UI elements): navbar, card, hero, form, etc.
  • Mixed: Can combine both in a single workflow

Autonomous Flow (⚠️ CONTINUOUS EXECUTION - DO NOT STOP):

  1. User triggers: /workflow:ui-design:explore-auto [params]
  2. Phase 5: Target confirmation → User confirms → IMMEDIATELY triggers Phase 7
  3. Phase 7 (style-extract) → Attach tasks → Execute → Collapse → Auto-continues to Phase 8
  4. Phase 8 (animation-extract, conditional):
    • IF should_extract_animation: Attach tasks → Execute → Collapse → Auto-continues to Phase 9
    • ELSE: Skip (use code import) → Auto-continues to Phase 9
  5. Phase 9 (layout-extract) → Attach tasks → Execute → Collapse → Auto-continues to Phase 10
  6. Phase 10 (ui-assembly)Attach tasks → Execute → Collapse → Workflow complete

Phase Transition Mechanism:

  • Phase 5 (User Interaction): User confirms targets → IMMEDIATELY executes Phase 7
  • Phase 7-10 (Autonomous): Skill execute ATTACHES tasks to current workflow
  • Task Execution: Orchestrator EXECUTES these attached tasks itself
  • Task Collapse: After tasks complete, collapse them into phase summary
  • Phase Transition: Automatically execute next phase after collapsing
  • No additional user interaction after Phase 5 confirmation

Auto-Continue Mechanism: TodoWrite tracks phase status with dynamic task attachment/collapse. After executing all attached tasks, you MUST immediately collapse them, restore phase summary, and execute the next phase. No user intervention required. The workflow is NOT complete until Phase 10 (UI assembly) finishes.

Read the full file on GitHub · 679 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 679 lines · 0 tokens per session scan C 19efc45fa068

Subscribe to this mod's changes

explore-auto is a command published in the GitHub repository catlog22/Claude-Code-Workflow (2,135 stars, last pushed 2mo ago), licensed MIT. It adds 27 tokens to every session and 7,384 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.