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 autonomous-ai/autonomous-workshop --skill product-designgit clone --depth 1 https://github.com/autonomous-ai/autonomous-workshopWrote 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/autonomous-ai/autonomous-workshop/product-design)<a href="https://agentmods.dev/skills/autonomous-ai/autonomous-workshop/product-design"><img src="https://agentmods.dev/badge/skills/autonomous-ai/autonomous-workshop/product-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/autonomous-ai/autonomous-workshop/product-design"><img src="https://agentmods.dev/badge/skills/autonomous-ai/autonomous-workshop/product-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 | $0.00203 | $0.04716 |
| Opus 5.5 | $0.00081 | $0.01886 |
| Sonnet 5.5 | $0.00041 | $0.00943 |
| Haiku 4.5 | $0.00020 | $0.00472 |
Grade A, and why
product-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 16d 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 — 342 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Product design — design the object before CAD builds it
Purpose
A reference image arrives with its design already made, and $image-to-cad
reads it. A prose request arrives without one: "a phone stand" fixes a category
and nothing else. Handed straight to $cad, the CAD turn becomes the designer
and builds the first shape that satisfies the words — a block with a slot, one
small fillet on every edge. No gate notices: every geometry check passes an
object nobody would want on their desk.
This skill is the designer the flow was missing. Work the way an industrial
designer does: understand what the object is for and who lives with it, study
how it is already made and which versions people find beautiful, set a design
direction before drawing anything, explore concepts and choose one on purpose,
and then critique the built model in rounds until it looks right. The
object must work, print and fit — and it must be beautiful. Beauty is not left
to whatever the first CAD pass produces; it is a requirement, designed in order
(big form to detail) and reviewed against written rules
(wiki show product-aesthetics).
You produce a document, not geometry. Fill templates/design_spec.md and
hand off to $cad; the design-review rounds then run on $cad's renders and are
recorded back in the spec.
Inside a Workshop product run resolve
CAD_SKILL_ROOT="$(workshop skills path)/cad" and
IMAGE_TO_CAD_SKILL_ROOT="$(workshop skills path)/image-to-cad", and read a
skills/<name>/... path in these pages under
"$(workshop skills path)/<name>/". Upstream's $stl-to-step and
$step-to-source are not Workshop skills; a supplied STL or STEP still does
not fire this one. Five things change there:
- It fires only while the design is open: a prose Wish with no sealed
Invent concept, Design Contract or reference image. A sealed concept or
contract is the design, so its survey, direction and concept selection are
never reopened here; an image to reproduce goes to
$image-to-cad. - The selected Inventor's
TASTE.mdgoverns creative judgment. The design direction works inside it, and the Wish's words stand where this page says the user's words do. There is no user to ask or to accept a finding. - Review rounds spend Make's own allowance. They run inside the round and review budget the host froze, never replace the Make reference's blind signature review, and stop when that budget does; a finding still open then is reported in the final response, not chased past it.
- The spec lives in the run's CAD project, beside the entries
$cadbuilds, not underoutput/. - Final verification is the one the Make reference names. Do not add
--freshin a product run: the host performs the isolated fresh rebuild.
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.
- 16d ago First seen · 342 lines · 203 tokens per session scan A 4d101af4d3d9
product-design is a skill published in the GitHub repository autonomous-ai/autonomous-workshop (21 stars, last pushed today), licensed Apache-2.0. It adds 203 tokens to every session and 4,716 once invoked, about $0.0008 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 2026-10-01.
Other skills, from other repositories
cocoindex-diagrams
This skill should be used when creating, editing, or reviewing inline SVG diagrams for the CocoIndex docs site (anything under docs/src/content/docs/). It encodes the component-based primitive system under docs/src/components/diagrams/, the palette + shape semantics, the preview-and-verify loop using headless Chrome…
cli-command
Design, implement, or review Composio CLI commands under ts/packages/cli using Effect, effect/cli, services, output conventions, configuration and environment variables, and local vendor references. Use for CLI command UX, command wiring, service changes, or CLI source edits. Do not use for CLI E2E-only work; use…
dashboard-design
Use this skill first when the user wants to design or plan a dashboard, especially Vizro dashboards. Enforces a 3-step workflow (requirements, layout, visualization) before implementation. Activate when the user asks to create, design, or plan a dashboard. For implementation, use the dashboard-build skill after…
dstack-prototyping
Use with the dstack skill for model-serving work when the image, serving command, resources, backend/fleet choice, or service behavior is not proven. Guides task-first prototyping on real hardware, choosing fleets/backends that can reuse idle instances and caches, checking vLLM/SGLang sources, and verifying the final…
designing-vizro-layouts
Use this skill when designing or building Vizro dashboard layouts — grid configuration, component sizing, filter/parameter placement, selector types, or container patterns. Activate when the user is creating wireframes, defining page structure, placing controls, or sizing charts.
ui-mobile
Mobile UI patterns - React Native, iOS/Android, touch targets.