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 tranfu-labs/tranfu-skills --skill visual-dna-systemgit clone --depth 1 https://github.com/tranfu-labs/tranfu-skillsWrote 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/tranfu-labs/tranfu-skills/visual-dna-system)<a href="https://agentmods.dev/skills/tranfu-labs/tranfu-skills/visual-dna-system"><img src="https://agentmods.dev/badge/skills/tranfu-labs/tranfu-skills/visual-dna-system/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/skills/tranfu-labs/tranfu-skills/visual-dna-system"><img src="https://agentmods.dev/badge/skills/tranfu-labs/tranfu-skills/visual-dna-system.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 | $0.00148 | $0.01960 |
| Opus 5 | $0.00074 | $0.00980 |
| Sonnet 5 | $0.00030 | $0.00392 |
| Haiku 4.5 | $0.00015 | $0.00196 |
Grade A, and why
visual-dna-system 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 9d 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Visual DNA System
Purpose
MUST extract transferable visual principles from a visual sample and convert them into a reusable abstract design system. MUST remove source brand identity, protected layouts, exact copy, and proprietary components. NEVER reuse source logos, brand names, exact layouts, exact copy, or proprietary components.
Inputs
Accept one or more of:
- Website URL, screenshot, Figma reference, image, HTML/CSS, app UI, deck, poster, card, dashboard, or other visual artifact.
- Optional user notes about what attracted them to the sample.
- Optional target industry, audience, elements to ignore or emphasize, abstraction level, and originality constraints.
If source access requires tools that are unavailable, ask for a screenshot, file, or visual description.
Workflow
CREATE A TODO LIST FOR THE TASKS BELOW:
- Confirm required references are readable:
references/output-schema.mdandreferences/originality-guardrails.md. If either file is missing or unreadable, reportBLOCKER reference-unavailable: <path>and stop. - Read the visual sample and user notes. If no usable sample exists, report
missing-visual-sample, ask for a URL, screenshot, file, or visual description, and stop. - Apply input guards before extraction:
- If the user asks only for final production, report
production-requested, return a short copyable handoff prompt for any available downstream HTML/design producer, and stop. - If visual details are too vague to infer, report
insufficient-visual-detail, ask one focused clarification question or request a better visual sample, and stop. - If the user asks for direct copying or source brand recreation, report
source-copy-request, NEVER produce that copied part, and continue with abstract extraction only. - If evidence is partial, report
partial-evidence, continue with lower confidence, and fillEvidence And Confidence/missing_evidence.
- If the user asks only for final production, report
- Classify the evidence: URL, screenshot, Figma, image, HTML/CSS, deck, poster, UI, or mixed source.
- If a URL is unreachable because of login, anti-bot protection, 404, or network failure, report
source-unreachable, ask for screenshots or HTML/CSS snippets, and stop. - If the sample is non-design content such as a natural photo, video frame, or random screenshot with no design intent, report
insufficient-design-signal, request a sample with clear design intent, and stop. - If multiple samples have conflicting styles, report
conflicting-style-samples, list the differences, and ask whether to synthesize one main DNA, output separate DNA systems, or choose one primary sample. If the user does not answer, output separate DNA systems and continue. - Extract visible design signals across essence, color, typography, layout, components, material, imagery, iconography, motion, and composition.
- Separate transferable principles from non-transferable identity markers.
- MUST remove source brand assets, logos, brand names, exact layouts, proprietary components, exact copy, and unique identity markers.
- Compose the Markdown
Visual DNA Design Systemusing the schema inreferences/output-schema.md. - Emit the JSON/tokens representation using the schema in
references/output-schema.md. - Compose a copyable downstream production prompt for any available downstream HTML/design producer.
- Apply the originality guardrails in
references/originality-guardrails.md. If any Self-Check answer is not a clear "No", return to steps 9-10 and rewrite the transferable/non-transferable split. Retry at most 2 times; if the gate still fails, reportguardrails-cannot-converge, omit disputed transfer items, keep only high-confidence abstract principles, list omitted items inOriginality Guardrails, and continue to step 15. - Deliver the named artifacts and end.
What ships with it
8 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.
- 9d ago First seen · 130 lines · 148 tokens per session scan A 4a583ddc923b
visual-dna-system is a skill published in the GitHub repository tranfu-labs/tranfu-skills (2 stars, last pushed 2d ago), licensed MIT. It adds 148 tokens to every session and 1,960 once invoked, about $0.0007 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 skills, from other repositories
webgl-holographic-foil
A self-contained WebGL2 hero: thin-film interference over a crushed-foil surface whose palette shifts with the viewing angle; move the cursor to tilt the film.
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
html-ppt-taste-brutalist
16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).
visual-ralph
Visual Ralph orchestration for frontend UI from generated references, static references, or live URL targets, using $ultragoal with built-in visual verdict and pixel-diff evidence until the implementation matches and leaves a reproducible design system.
accessibility
Consolidated accessibility skill entrypoint for WCAG 2.2, ARIA Authoring Practices, cognitive accessibility, Section 508, EN 301 549, design intent verification, and the Accessibility Planner workflow.
make-resume
A Chinese-language tool for creating editable HTML resumes that can be changed in a browser and printed to PDF. It uses available resume templates when they are installed and otherwise provides a simpler fallback.