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.
npx skills add lukedj78/dev-flow --skill image-to-design-mdgit clone --depth 1 https://github.com/lukedj78/dev-flowWrote 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/skills/lukedj78/dev-flow/image-to-design-md)<a href="https://agentmods.dev/skills/lukedj78/dev-flow/image-to-design-md"><img src="https://agentmods.dev/badge/skills/lukedj78/dev-flow/image-to-design-md.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 166 Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.Fix: Remove instructions that suppress warnings, disclaimers, or ethical commentary. Let the agent surface safety-relevant caveats to the user.
- medium analysis-evasion · line 1 Suspicious Unicode normalization or mixed-script contentFix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
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 | $0.00190 | $0.02993 |
| Opus 5 | $0.00095 | $0.01496 |
| Sonnet 5 | $0.00038 | $0.00599 |
| Haiku 4.5 | $0.00019 | $0.00299 |
Grade A, and why
image-to-design-md 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 7d 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 — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Image(s) → DESIGN.md
Take 1 or more raster images (PNG / JPG) and produce a DESIGN.md that conforms to the Google design.md spec. The same workflow used in figma-to-design-md Path C (manual fallback) is the default here, since raster is all we have.
When this skill applies
Trigger on any of:
- The user pastes 1+ image paths and asks for a design system / DESIGN.md.
- The user describes images they have (Pinterest pins, Dribbble shots, competitor screenshots, hand-drawn mockups, scanned printouts).
- An explicit request: "estrai design da queste immagini", "make a DESIGN.md from this image", "build a design system from this Figma export PNG".
- The orchestrator (
dev-flow) routes here from phaseprd_draftedoremptywhen the user has no Figma URL but has images.
If the user has BOTH a Figma URL AND images, prefer figma-to-design-md (more precise — Variables panel beats vision LLM) and pass the images to screenshot-to-page later.
What you produce
In dev-flow mode: <root>/.workflow/DESIGN.md + <root>/.workflow/screenshots/<slug>.png for each input image.
In standalone mode: <project-root>/DESIGN.md + <project-root>/screenshots/.
The DESIGN.md follows the same 8-section structure as figma-to-design-md produces — same spec — see references/spec.md for the canonical Google design.md spec.
Dev-flow contract
This skill participates in the dev-flow workflow. See references/contracts.md for the canonical schema. Key facts:
- Output goes into
<root>/.workflow/:DESIGN.md(the main artifact)screenshots/(input images copied + slugged)
- State is updated by setting
meta.json#phase = "design_extracted"(only if the current phase is earlier in the enum), refreshingupdated_at, and appending ahistoryentry. - Standalone mode (no
.workflow/present) is supported — fall back to writing at the user-specified project root.
Prerequisites
- Vision-capable LLM. This skill leans on the model's vision capability for typography identification, component recognition, and layout inference. Claude Opus/Sonnet/Haiku 4.5+ all qualify; ChatGPT-4o and Gemini also work in their respective harnesses.
- Python 3.10+ with Pillow + numpy + scikit-learn (only for
quantize_palette.py).
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 7d ago First seen · 177 lines · 190 tokens per session scan A f9903f71b1fe
image-to-design-md is a skill published in the GitHub repository lukedj78/dev-flow (6 stars, last pushed yesterday), licensed MIT. It adds 190 tokens to every session and 2,993 once invoked, about $0.0010 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 2026-08-31.
Other skills, from other repositories
mcp-host-styling-integration
Integrates MCP App UI with host theming system. Applies host CSS variables, handles onhostcontextchanged, safe area insets, display mode detection, and fullscreen configuration.
moai-design-tools
Design tool integration specialist covering Figma MCP, Pencil renderer, and Pencil-to-code export. Use when fetching design context from Figma, rendering Pencil designs, or exporting to React/Tailwind code.
taiyi-ui-design
A design-planning guide for describing how an application's user interface should look and behave. It produces a UI-DESIGN.md document covering layouts, components, interactions, accessibility, and error states.
taiyi-diagram-render
A tool for rendering existing C4 or architecture Mermaid diagrams as SVG or PNG images. C4 is a way to show software systems at different levels, while Mermaid is a text format for diagrams.
taiyi-diagram-pipeline
A pipeline that creates and updates architecture diagrams in three stages: a C4 model from the code, an engineering architecture diagram, and an exported PNG image. C4 is a way to describe software architecture at several levels of detail.
ux-design
Create comprehensive UX design with user flows, wireframes, and design systems.