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 nguyenthienthanh/aura-frog --skill design-vision-loopgit clone --depth 1 https://github.com/nguyenthienthanh/aura-frogWrote 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/nguyenthienthanh/aura-frog/design-vision-loop)<a href="https://agentmods.dev/skills/nguyenthienthanh/aura-frog/design-vision-loop"><img src="https://agentmods.dev/badge/skills/nguyenthienthanh/aura-frog/design-vision-loop.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
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 →
- medium Agent Snooping · line 93 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 94 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00080 | $0.01258 |
| Opus 5 | $0.00040 | $0.00629 |
| Sonnet 5 | $0.00016 | $0.00252 |
| Haiku 4.5 | $0.00008 | $0.00126 |
Grade A, and why
design-vision-loop 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 8d 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI-consumed reference. This skill verifies rendered UI; it does not generate taste (that's
frontend-aesthetics) or enforce statically (that's thedesign-conformancehook +theme-consistencyrule). Close the loop: build → see → critique → fix.
Design Vision Loop
Open-loop text guidance is why "correct" UI still looks off — nobody looked at it. This skill adds the eyes: render → screenshot → critique → iterate, the pattern Anthropic endorses for UI agents (Building agents with the Claude Agent SDK: "visual feedback… can be helpful"; computer-use docs: "after each step, take a screenshot and carefully evaluate… only when you confirm a step was executed correctly should you move on").
Precondition
The UI must be renderable: a running dev server URL/route, or a static HTML artifact. If nothing renders (component with no harness), say so and either spin up a preview or fall back to static review — don't pretend to have looked.
The loop (max 3 iterations by default)
loop[4]{step,action,tools}:
1,Render + capture,"navigate → resize each viewport → screenshot; emulate dark mode, screenshot again"
2,Tier-1 deterministic gates,"run the design-conformance check + read console errors — cheap, run FIRST"
3,Tier-2 vision critique,"read the screenshots, score against references/critique-rubric.md"
4,Verdict,"PASS → done · defects → fix → re-enter step 1 (until pass or budget hit)"
Two-tier verification is deliberate (Anthropic ranks rules-based feedback above LLM-as-judge): the deterministic gate is the strong signal and it's nearly free, so run it before spending a vision call. Only screenshot-critique what already passes the lint-level gate.
Step 1 — capture (Playwright MCP)
browser_navigateto the route → for each viewport inreferences/viewport-matrix.md(browser_resize375 → 768 → 1440) →browser_take_screenshot.- Dark mode: emulate
prefers-color-scheme: dark(viabrowser_evaluate/ an emulation call) and re-screenshot at the primary viewport. If the design system declares no dark mode, skip and note it. - Save PNGs under
.claude/workflow/vision/<route>-<viewport>[-dark].png.
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.
- 8d ago First seen · 95 lines · 80 tokens per session scan A 08d20c8d6b8b
design-vision-loop is a skill published in the GitHub repository nguyenthienthanh/aura-frog (24 stars, last pushed 3d ago), licensed MIT. It adds 80 tokens to every session and 1,258 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
gsd-sketch
Sketch UI/design ideas with throwaway HTML mockups, or propose what to sketch next (frontier mode).
gsd-ui-phase
Generate UI design contract (UI-SPEC.md) for frontend phases.
ux-design
Guided, section-by-section UX spec authoring for a screen, flow, or HUD. Reads game concept, player journey, and relevant GDDs to provide context-aware design guidance. Produces ux-spec.md (per screen/flow) or hud-design.md using the studio templates.
recipe-front-design
Execute from repository evidence through applicable UI Spec and optional ADR decisions to complete frontend Design Doc approval.
design-an-interface
Produce 3+ radically different designs for a module, API, or interface, compare them in prose, and synthesize a recommendation. Use when asked to "design an interface", "shape this API", "design it twice", "explore module boundaries", or when planning a new deep module and the first idea is unlikely to be the best.…
openspec-plus-apply
MANDATORY skill that activates whenever the OpenSpec apply phase begins. Triggers: /opsx-apply runs, the openspec-apply-change vanilla skill is referenced or active, openspec instructions apply is invoked, or the user asks to implement, apply, execute, or build out an OpenSpec change ('implement the change', 'apply…