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.
git clone --depth 1 https://github.com/catlog22/Claude-Code-WorkflowWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/catlog22/claude-code-workflow/style-extract)<a href="https://agentmods.dev/commands/catlog22/claude-code-workflow/style-extract"><img src="https://agentmods.dev/badge/commands/catlog22/claude-code-workflow/style-extract/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/catlog22/claude-code-workflow/style-extract"><img src="https://agentmods.dev/badge/commands/catlog22/claude-code-workflow/style-extract.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00022 | $0.06858 |
| Opus 5 | $0.00011 | $0.03429 |
| Sonnet 5 | $0.00004 | $0.01372 |
| Haiku 4.5 | $0.00002 | $0.00686 |
Grade A, and why
style-extract scanned grade A with 0 findings 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 5d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 778 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auto Mode
When --yes or -y: Skip all clarification questions, use AI-inferred design decisions.
Style Extraction Command
Overview
Extract design style from reference images or text prompts using Claude's built-in analysis. Supports two modes:
- Exploration Mode (default): Generate multiple contrasting design variants
- Refinement Mode (
--refine): Refine a single existing design through detailed adjustments
Strategy: AI-Driven Design Space Exploration
- Claude-Native: 100% Claude analysis, no external tools
- Direct Output: Complete design systems (design-tokens.json)
- Flexible Input: Images, text prompts, or both (hybrid mode)
- Dual Mode: Exploration (multiple contrasting variants) or Refinement (single design fine-tuning)
- Production-Ready: WCAG AA compliant, OKLCH colors, semantic naming
Execution Process
Input Parsing:
├─ Parse flags: --design-id, --session, --images, --prompt, --variants, --interactive, --refine
└─ Decision (mode detection):
├─ --refine flag → Refinement Mode (variants_count = 1)
└─ No --refine → Exploration Mode (variants_count = --variants OR 3)
Phase 0: Setup & Input Validation
├─ Step 1: Detect input mode, extraction mode & base path
├─ Step 2: Load inputs
└─ Step 3: Memory check (skip if exists)
Phase 1: Design Direction/Refinement Options Generation
├─ Step 1: Load project context
├─ Step 2: Generate options (Agent Task 1)
│ └─ Decision:
│ ├─ Exploration Mode → Generate contrasting design directions
│ └─ Refinement Mode → Generate refinement options
└─ Step 3: Verify options file created
Phase 1.5: User Confirmation (Optional)
└─ Decision (--interactive flag):
├─ --interactive present → Present options, capture selection
└─ No --interactive → Skip to Phase 2
Phase 2: Design System Generation
├─ Step 1: Load user selection or default to all
├─ Step 2: Create output directories
└─ Step 3: Launch agent tasks (parallel)
Phase 3: Verify Output
├─ Step 1: Check files created
└─ Step 2: Verify file sizes
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.
- 5d ago First seen · 778 lines · 0 tokens per session scan A 3ac83e7faf1b
style-extract is a command published in the GitHub repository catlog22/Claude-Code-Workflow (2,131 stars, last pushed 2mo ago), licensed MIT. It adds 22 tokens to every session and 6,858 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
multi-frontend
Run a frontend-focused multi-model workflow for components, layouts, animation, and UI polish.
create-excalidraw-diagram
description: "Create system architecture or technical diagram (Excalidraw) (Use any time you need a diagram)".
bmad-ux
Capture the user's UX vision in two documents: DESIGN.md for how the product looks and EXPERIENCE.md for how it behaves. Use when the user says "lets create UX design" or "create UX specifications" or "help me plan the UX".
bmad-cis-design-thinking
Guide human-centered design processes using empathy-driven methodologies. Use when the user says "lets run design thinking" or "I want to apply design thinking".
my_brand
Interactive brand identity creation with archetypes, colors, typography, and voice guidelines. Use when user mentions "brand", "colors", "typography", "brand.md", "design system", or "visual identity".
verify-claim
Walk a claim through the SIFT method (Stop, Investigate, Find better coverage, Trace).