design-graph: Skill for Claude Code

.claude/skills/design-graph-authoring/SKILL.md

design-graph-authoring is a skill for Claude Code from noemica-io/design-graph. It costs 145 tokens per session (3,643 once invoked), scanned A, original, MIT.

A guide for creating multiple, genuinely different HTML page variants in Design Graph, a canvas for viewing and comparing those variants.

In plain words
What is it for?
Use it to generate variant pages, add them as nodes and connections in the graph data, and write useful descriptions for comparison.
Why use it?
It keeps the canvas updated as each page is completed, so reviewers can see and compare results without waiting for every variant.

Skill for Claude Code

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

This is noemica-io/design-graph's own configuration. It tells Claude Code how to work on design-graph 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 design-graph configures →

Reuse

Borrowing it

Nothing to install: this file belongs to noemica-io/design-graph. 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/noemica-io/design-graph/main/.claude/skills/design-graph-authoring/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/noemica-io/design-graph

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-graph-authoring

README.md
[![agentmods](https://agentmods.dev/badge/skills/noemica-io/design-graph/design-graph-authoring/github.svg)](https://agentmods.dev/skills/noemica-io/design-graph/design-graph-authoring)
Your own site
<a href="https://agentmods.dev/skills/noemica-io/design-graph/design-graph-authoring"><img src="https://agentmods.dev/badge/skills/noemica-io/design-graph/design-graph-authoring/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-graph-authoring

Your own site · 80×15
<a href="https://agentmods.dev/skills/noemica-io/design-graph/design-graph-authoring"><img src="https://agentmods.dev/badge/skills/noemica-io/design-graph/design-graph-authoring.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 145 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,643 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 $0.00145 $0.03643
Opus 5 $0.00072 $0.01821
Sonnet 5 $0.00029 $0.00729
Haiku 4.5 $0.00015 $0.00364

Measured 10d ago against content hash a8b6713d4de4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

design-graph-authoring 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 10d 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.

.claude/skills/design-graph-authoring/SKILL.md · 321 lines

How it starts

The opening of the file, as written. The whole thing — 321 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Design Graph Authoring

Design Graph is a canvas for reviewing many HTML page variants side by side. The canvas viewer doesn't generate anything — it just renders graph-data.js and serves the files in pages/. You are the generator. This skill is how you do that well.

The single most important rule: write eagerly

The moment a sub-agent reports done, append its node + edges to graph-data.js immediately. Do not wait for the batch. Do not collect all returns and write them at the end. The user is watching the canvas in live mode (?live is the default) and expects each card to pop into view the second its HTML is on disk.

Why this matters: the perceived speed of the whole session is set by how quickly the first card appears, not by when the last one finishes. If you batch writes, the user stares at an empty canvas for 60+ seconds even though four agents finished in 15. Eager writing makes the tool feel alive.

The write itself is tiny — one {node} + one or two {edges} appended to a JSON-like array. It takes a single Edit call. There is no reason to batch. Even if two sub-agents return within the same second, process them sequentially in arrival order; the second Edit runs 20ms after the first.

Rule of thumb: from "sub-agent done" notification to graph-data.js updated should be under 5 seconds. If you catch yourself saying "let me wait for the other three to finish and then I'll add them all," stop and write the one you have.

The core loop, compressed

  1. Read the parent node + every existing sibling that shares the target label.
  2. Decide on N creative directions that are genuinely different (not rewordings).
  3. Spawn N background sub-agents — one per direction — each returning its HTML + the suggested node/edge entries.
  4. As each sub-agent reports done, append its entries to graph-data.js immediately. This is non-negotiable — see the eager-write rule above.
  5. Tell the user which nodes to open in the viewer.

Everything below is elaboration on that loop.

Read the full file on GitHub · 321 lines

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. 10d ago First seen · 321 lines · 145 tokens per session scan A a8b6713d4de4

Subscribe to this mod's changes

design-graph-authoring is a skill published in the GitHub repository noemica-io/design-graph (17 stars, last pushed 4mo ago), licensed MIT. It adds 145 tokens to every session and 3,643 once invoked, about $0.0007 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.