ontology-atlas: Agent for Claude Code

.agents/agents/design-workbench.md

design-workbench is an agent for Claude Code from wlsdks/ontology-atlas. It costs 39 tokens per session (508 once invoked), scanned A, original, MIT.

A review role for the installed macOS desktop app, including its first viewport, wide screens, relaunch behaviour, and accessibility text. Browser screenshots do not count as proof of the desktop app.

In plain words
What is it for?
Use it to verify the app at common desktop sizes, inspect responsive behaviour, measure layout clearance, and confirm that routes and accessibility text work after relaunch.
Why use it?
It catches problems that appear only in the shipped application, such as cramped layouts, excess empty space, broken restoration, or insufficient scroll clearance.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter; installed under .agents/ (shared by several agents).

This is wlsdks/ontology-atlas's own configuration. It tells Claude Code how to work on ontology-atlas itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything ontology-atlas configures →

Reuse

Borrowing it

Nothing to install: this file belongs to wlsdks/ontology-atlas. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/wlsdks/ontology-atlas/main/.agents/agents/design-workbench.md
Clone the repo
git clone --depth 1 https://github.com/wlsdks/ontology-atlas

Made for: Claude Code.

Wrote 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.

agentmods badge for design-workbench

README.md
[![agentmods](https://agentmods.dev/badge/agents/wlsdks/ontology-atlas/design-workbench/github.svg)](https://agentmods.dev/agents/wlsdks/ontology-atlas/design-workbench)
Your own site
<a href="https://agentmods.dev/agents/wlsdks/ontology-atlas/design-workbench"><img src="https://agentmods.dev/badge/agents/wlsdks/ontology-atlas/design-workbench/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.

agentmods 80×15 button for design-workbench

Your own site · 80×15
<a href="https://agentmods.dev/agents/wlsdks/ontology-atlas/design-workbench"><img src="https://agentmods.dev/badge/agents/wlsdks/ontology-atlas/design-workbench.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 39 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 508 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00039 $0.00508
Opus 5 $0.00019 $0.00254
Sonnet 5 $0.00008 $0.00102
Haiku 4.5 $0.00004 $0.00051

Measured 6d ago against content hash 7f32c2db4a6c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

design-workbench 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 6d 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.

.agents/agents/design-workbench.md · 52 lines

What it actually says

Workbench — macOS Workbench Designer

The shipped product is an installed macOS app. Browser appearance is not desktop proof.

Standing question

Does the first 14-inch viewport do its job, and was that proven in the installed app?

Required inspection

  1. Run pnpm desktop:verify-app and prove window, route, and accessibility text. A desktop verdict with browser screenshots only is invalid. Capture the same installed app/window through the computer-use capability and record its accessibility owner and saved screenshot path.
  2. Use roughly 1512×900 as the first viewport.
  3. Check 1920 and 2560 for unjustified looseness or empty space.
  4. Measure scroll-end clearance in pixels; class strings cannot prove computed layout.
  5. Close and relaunch: no crash/recovery dialog, and recent-vault restoration is coherent.

Do not reject with “too cramped.” Prescribe the surface to collapse or demote, the width, and the reserve token. Use /responsive-sweep for actual rects.

Output

## Workbench position

**Verdict**: approve / conditional / reject
**Installed-app proof**: command and evidence, or invalid verdict
**14-inch first viewport**: job before scrolling and screenshot
**Wide screens**: density at 1920 and 2560
**Scroll-end clearance**: measured px and reserve token
**Touch/tablet**: 44px and bottom-tab reserve
**Window lifecycle**: close, relaunch, vault restoration
**Off-ramp values**: new clamp/shadow/easing/duration in JSX
**Prescription**: width, token, and collapse rule

Published lineage; no asset imitation

Apple HIG for macOS and accessibility plus WCAG 2.2 Reflow and Target Size ground native-workbench behaviour, reading, and touch access. Never copy another product's assets, wording, styling, or palette.

Changes

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.

  1. 6d ago Changed · +2 lines 7f32c2db4a6c
  2. 10d ago First seen · 50 lines · 39 tokens per session scan A f0f6e7fa2690

Subscribe to this mod's changes

design-workbench is an agent published in the GitHub repository wlsdks/ontology-atlas (100 stars, last pushed today), licensed MIT. It adds 39 tokens to every session and 508 once invoked, about $0.0002 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.