agentmods.dev

autonomous-workshop: Skill for Claude Code

.claude/skills/design-a-toy/SKILL.md

design-a-toy is a skill for Claude Code from autonomous-ai/autonomous-workshop. It costs 138 tokens per session (20,447 once invoked), scanned A, original, Apache-2.0.

A design workflow for defining a complete Workshop toy before its build run starts. It questions the idea, freezes a specification, creates reference images for distinct shapes, and produces the command that starts the Workshop process.

In plain words
What is it for?
Use it when starting a new toy or correcting a previous run that no longer matches the desired design.
Why use it?
It prevents the build run from inventing missing design decisions or drifting away from the intended toy.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: reads .claude/ paths; mentions subagents.

This is autonomous-ai/autonomous-workshop's own configuration. It tells Claude Code how to work on autonomous-workshop 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 autonomous-workshop configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is uv run python src/workshop/make/skills/print-details/scripts/print_sheets.py \.

not rated 24repo +1 changed yesterday A scan 138 tokens original Apache-2.0
View source ↗ autonomous-ai/autonomous-workshop
Reuse

Borrowing it

Nothing to install: this file belongs to autonomous-ai/autonomous-workshop. 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/autonomous-ai/autonomous-workshop/main/.claude/skills/design-a-toy/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/autonomous-ai/autonomous-workshop

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-a-toy

README.md
[![agentmods](https://agentmods.dev/badge/skills/autonomous-ai/autonomous-workshop/design-a-toy/github.svg)](https://agentmods.dev/skills/autonomous-ai/autonomous-workshop/design-a-toy)
Your own site
<a href="https://agentmods.dev/skills/autonomous-ai/autonomous-workshop/design-a-toy"><img src="https://agentmods.dev/badge/skills/autonomous-ai/autonomous-workshop/design-a-toy/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-a-toy

Your own site · 80×15
<a href="https://agentmods.dev/skills/autonomous-ai/autonomous-workshop/design-a-toy"><img src="https://agentmods.dev/badge/skills/autonomous-ai/autonomous-workshop/design-a-toy.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 138 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 20,447 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 $0.00138 $0.20447
Opus 5.5 $0.00055 $0.08179
Sonnet 5.5 $0.00028 $0.04089
Haiku 4.5 $0.00014 $0.02045

Measured yesterday against content hash 096d6398ecac, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

design-a-toy 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 yesterday.

The scan reads SKILL.md. This mod also ships 7 executable files (scripts/camera_composition.py, scripts/contract_lint.py, scripts/gen_image.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

.claude/skills/design-a-toy/SKILL.md · 1,393 lines

How it starts

The opening of the file, as written. The whole thing — 1,393 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Design a toy before the run

This skill does the creative work outside Workshop, where the human watches every step, and hands Workshop a finished design.

Workshop's Spark route is ("make", "release"): the Wish goes in, Make builds, Release publishes. Nothing in the run verifies that the Manager consulted the Inventor or passed it the Wish verbatim — the Inventor subagent's instructions carry only its Taste. The answer is not to police the run. It is to leave nothing for the run to invent.

The output of this skill is one CONTRACT.md, handed to build-a-toy.

The human reviews pictures, not the contract. They answer the design questions in Stage 1 and approve the finished set of images in Stage 4. They never read or approve the design-contract block or a contract diff. Keeping the contract and the images in agreement is this skill's job, done by the Contract Proof and measurement in Stages 3a and 3b, so that approving the images is the same as approving the contract.

Lessons

LESSONS.md holds what earlier designs and builds taught, with its Stage 3 entries in LESSONS-REFERENCE-IMAGES.md. Read its section for each stage before working that stage. It never overrides this file, the Taste or the human's answer.

Who designs

Read the Inventor's TASTE.md from inventors/<id>/TASTE.md and work as that Inventor throughout. The Taste is the constitution: it decides what is in scope, what wins a trade-off, and what gets rejected. Do not paraphrase it into something more agreeable. If the idea the human wants is outside the Taste, say so plainly and name which clause it fails — do not quietly widen the Taste to fit.

There is no default Inventor. Read the description line of every inventors/*/TASTE.md, propose the one or two whose scope fits the idea, name the clause that makes each fit, and let the human choose. A board-game Inventor is not a safe fallback for a vehicle, a creature or a mechanism.

Read the full file on GitHub · 1,393 lines

Files

What ships with it

13 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.

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. yesterday Changed · +1 lines 096d6398ecac
  2. 3d ago Changed · +91 lines · +4 tokens per session 7e266cacdc31
  3. 4d ago Changed · +2 lines 11b76b007d1d
  4. 6d ago Changed · +2 lines 73e4f4cc1dfa
  5. 7d ago Changed · +333 lines · +11 tokens per session a02555c8c4b5
  6. 13d ago Changed · +74 lines 7adbf0952901
  7. 16d ago Changed · +63 lines a59b7cf274ce
  8. 20d ago Changed · +4 lines 3573c103e0aa
  9. 22d ago Changed · +44 lines 9637cb47c444
  10. 28d ago First seen · 779 lines · 123 tokens per session scan A 8f44bc82195a

Subscribe to this mod's changes

design-a-toy is a skill published in the GitHub repository autonomous-ai/autonomous-workshop (24 stars, last pushed yesterday), licensed Apache-2.0. It adds 138 tokens to every session and 20,447 once invoked, about $0.0006 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 2026-09-17.

3d-printingGame developmentPlanningPythonaiai-agentsgenerative-designtoys
Related

design-a-toy compared with skills from other repositories

design-a-toy next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
design-a-toyautonomous-ai/autonomous-workshopSkill24yesterdayApache-2.0A138Claude Code
pygame-coregamedev-skills/awesome-gamedev-agent-skillsSkill6.9k9d agoApache-2.0A85Claude Code
vibeuekevinpbuckley/VibeUESkill8122d agoMITA82Claude Code, Codex
demo-1-terminal-arcadedecodingai-magazine/building-a-coding-agent-from-scratch-courseSkill1.1k13d agoApache-2.0A46Claude Code, Codex
glyphs-mcp-developmentthierryc/Glyphs-mcpSkill595d agoApache-2.0A59Codex
game-ai-patternsHermeticOrmus/claude-code-game-developmentSkill781mo agoMITA58Claude Code
animation-patternsHermeticOrmus/claude-code-game-developmentSkill781mo agoMITA64Claude Code