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 agentlas-ai/agentlas-desktop --skill hig-reviewgit clone --depth 1 https://github.com/agentlas-ai/agentlas-desktopWrote 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/agentlas-ai/agentlas-desktop/hig-review)<a href="https://agentmods.dev/skills/agentlas-ai/agentlas-desktop/hig-review"><img src="https://agentmods.dev/badge/skills/agentlas-ai/agentlas-desktop/hig-review/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/agentlas-ai/agentlas-desktop/hig-review"><img src="https://agentmods.dev/badge/skills/agentlas-ai/agentlas-desktop/hig-review.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 | $0.00065 | $0.00948 |
| Opus 5.5 | $0.00026 | $0.00379 |
| Sonnet 5.5 | $0.00013 | $0.00190 |
| Haiku 4.5 | $0.00006 | $0.00095 |
Grade A, and why
hig-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Purpose
Produces a grounded design review: each finding names the guideline that backs it, carries a severity that reflects consequence, and lands as a concrete change in the user's framework — not Apple's. Also runs in improvement mode, where the review becomes a sequenced work plan.
Preconditions
- The surface under review is identified: a URL, a running app, a screenshot, a mockup, or code.
- The target platform and stack are known or inferable. A wrong platform guess invalidates the review.
- Guideline text is reachable — either the local cache (
references/apple-hig/.cache/pages/) or the network.$hig-sourcesexplains the cache contract and how to refill it. - Adhere to
$hig-review-protocoland$critical-overrides.
Steps
- Frame it. Establish surface, platform/stack, product job, and what the user wants (full audit, one concern, or improvements). Ask only what cannot be inferred.
- Get evidence. For a live URL or app, capture the screens with
@agentlas-browseror@playwright— the flow, not one screen. For code, read the components that render it. Never review from a verbal description when the real thing is reachable. - Route to guidelines. Open
$hig-lookupand pick 3–8 pages:accessibility,color,layout, andtypographyalways, plus the pages matching what is actually on screen. Read the cached page; if absent, fetch its DocC JSON as the routing table describes. - Audit. Work
$hig-checklistin order — accessibility, platform conventions, layout, visual system, interaction states, content, privacy. Measure what can be measured (contrast, target size, text scaling) instead of estimating it. - Translate. Convert every Apple noun into the user's stack with
$hig-platform-translation, and drop the rules that do not bind their platform rather than reporting them as violations. For glass, blur, or vibrancy work, apply$liquid-glass. - Grade and write. Assign severity by consequence and produce the report shape defined in
$hig-review-protocol, including what already works and the limits of the review. - Improvement mode (when asked to fix, not just judge). Rank findings by severity against effort, propose the concrete change for each, then sequence: accessibility first, platform conventions next, visual and polish last. Apply the changes only when the user asks for that.
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.
- 1mo ago First seen · 62 lines · 65 tokens per session scan A 04a9b7daeee2
hig-review is a skill published in the GitHub repository agentlas-ai/agentlas-desktop (7 stars, last pushed yesterday), licensed Apache-2.0. It adds 65 tokens to every session and 948 once invoked, about $0.0003 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 2026-09-04.
Other skills, from other repositories
architecture-diagram
How to draw and maintain the architecture diagram Ensemblr keeps for a workspace — ensemblrgetarchitecturediagram to fetch the stored one (it answers with null when nobody has drawn this workspace yet), the IR shape, how to derive it from the codebase, and ensemblrupdatearchitecturediagram to store it. Read this when…
winui-design
Read before writing or changing WinUI 3 XAML: choosing controls, pages and navigation, layout, Fluent Design, theming, accessibility, x:Bind, real samples, and checking that APIs exist with winapp find-api. Not for WPF or WinForms UI, even if the app uses the Windows App SDK.
openscreen-design
Use this skill to generate well-branded interfaces and assets for OpenScreen (an AI-native screen-recording studio & video editor), either for production or throwaway prototypes/mocks/etc. Contains essential design guidelines, colors, type, fonts, assets, and UI kit components for prototyping.
figma-generate-design
Use this skill alongside figma-use when the task involves translating an application page, view, or multi-section layout into Figma. Triggers: 'write to Figma', 'create in Figma from code', 'push page to Figma', 'take this app/page and build it in Figma', 'create a screen', 'build a landing page in Figma', 'update the…
edit-figma-design
Create or update Figma designs directly from a written product or UI description using the Figma MCP authoring tools. Use when the user wants a mockup, wireframe, screen, component, flow, or concept designed in Figma from text, or wants to iterate on an existing Figma file from textual feedback. Despite the name, this…
screen-annotation
Point at things on the screen the user is showing you.