Borrowing it
Nothing to install: this file belongs to Lengros/bennu. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Lengros/bennu/main/.claude/skills/design/SKILL.mdgit clone --depth 1 https://github.com/Lengros/bennuWrote 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/lengros/bennu/design)<a href="https://agentmods.dev/skills/lengros/bennu/design"><img src="https://agentmods.dev/badge/skills/lengros/bennu/design/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/lengros/bennu/design"><img src="https://agentmods.dev/badge/skills/lengros/bennu/design.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 | $0.00229 | $0.04049 |
| Opus 5.5 | $0.00092 | $0.01620 |
| Sonnet 5.5 | $0.00046 | $0.00810 |
| Haiku 4.5 | $0.00023 | $0.00405 |
Grade A, and why
design 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 2mo 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 — 261 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/design — commit to an art direction, then render and LOOK
This skill makes visual artifacts that win hearts, not "tasteful default" ones. It owns the part the generic loop keeps fumbling: the model is beautiful-by-default, which means bland-by-default. Asked to "make it nice" with no art direction, it regresses to the median of "tasteful product page" in its training — and that median is a specific look.
Core principle: Beauty that wins hearts is not "try harder" — it's a committed, named direction chosen BEFORE a line of HTML, an explicit ban on the default attractor, and a render→look→critique loop so the agent actually SEES what it made. Remove those three blockers and the same model that produced bland produces distinctive.
This is the identity counterpart to /diagram (which owns structure). For a
diagram, the safe-polished default — rounded cards, hairline borders, dotted bg, one
accent — is correct. For a hero/landing, that same default is the trap.
⛔ The default attractor — name it so you can escape it
Absent a committed direction, the model reaches for this exact combination. It is competent and forgettable. Recognize it and refuse it unless it is the chosen direction:
| Tell | The default (BAN unless chosen) |
|---|---|
| Background | warm cream #f4f6f1 / off-white, with faint radial-gradient "orb" blobs + dot-grid texture |
| Accent palette | sage green #1f6f54 + clay/terracotta #a25c1d (the "Claude/Anthropic" pairing) |
| Display type | Fraunces (or another opsz serif) for headings + Inter body |
| Cards | border-radius:18–22px, hairline #d9e0da border, soft 0 8px 26px rgba(...,.07) shadow |
| Mood | calm, editorial-SaaS, "thoughtful startup", safe |
When you catch your draft using ≥2 of these rows and nobody asked for that look, you are in the attractor. Stop and commit a real direction.
Worked exemplar — the gap this skill closes. Same content, two outcomes. Bland (default attractor):
--bg:#f4f6f1; --green:#1f6f54; --clay:#a25c1d, Fraunces headings, radius-18 cards, radial orbs. Reads "competent, generic, like every Claude page." Distinctive (committed direction — "cyberpunk control room"):--bg:#07090dnear-black, neon--cyan:#2ef7ff/--magenta:#ff3df2with color roles (cyan=system, magenta=AI),border-radius:0,clip-pathcorner notches, layered grid+scanline background, mono technical labels, a one-word glitch. Reads "mission systems online." The win came from the committed package, not a better model. The full package lives instyles/cyberpunk.md.
What ships with it
2 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.
- 2mo ago First seen · 261 lines · 229 tokens per session scan A d029b524498f
design is a skill published in the GitHub repository Lengros/bennu (2 stars, last pushed 3mo ago), licensed MIT. It adds 229 tokens to every session and 4,049 once invoked, about $0.0009 per session on Opus 5.5. 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
anatomy
Explain an idea as an interactive isometric machine figure.
claude-design
Design one-off HTML artifacts (landing, deck, prototype).
ip-as-logo
Design minimal cute IP mascot marks readable at 32px.
tldraw-offline
Drive and script tldraw offline canvases with an agent.
sketch
Throwaway HTML mockups: 2-3 design variants to compare.
baoyu-infographic
An infographic generator that turns supplied content into a visual summary using different information layouts and visual styles. It supports standard and custom image proportions and multiple languages.