Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add sdsrss/claudemd/plugin install claudemdWrote 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/commands/sdsrss/claudemd/claudemd-design-adopt)<a href="https://agentmods.dev/commands/sdsrss/claudemd/claudemd-design-adopt"><img src="https://agentmods.dev/badge/commands/sdsrss/claudemd/claudemd-design-adopt/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/commands/sdsrss/claudemd/claudemd-design-adopt"><img src="https://agentmods.dev/badge/commands/sdsrss/claudemd/claudemd-design-adopt.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 | $0.00145 | $0.01315 |
| Opus 5 | $0.00072 | $0.00658 |
| Sonnet 5 | $0.00029 | $0.00263 |
| Haiku 4.5 | $0.00015 | $0.00131 |
Grade A, and why
claudemd-design-adopt 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 2d 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Usage: /claudemd-design-adopt | /claudemd-design-adopt check | /claudemd-design-adopt remove
Step 0 — detect (deterministic)
Run: node ${CLAUDE_PLUGIN_ROOT}/scripts/design-detect.js --json (add --cwd=PATH to target another dir). The detector is stateless — it reads the filesystem and prints a verdict, writing nothing.
If --cwd is passed and names something that is not an existing directory (or is empty), the detector exits 2 with a stderr line instead of printing a verdict. That is not a finding about the project — report the bad path and stop; do not relay it as no-ui.
no-ui/ui-no-tokens→ report one line (nothing to adopt: <verdict>) and STOP. Never create DESIGN.md for a token-less project — generation would require inventing values, which this command forbids.configured→ report "already wired" + the DESIGN.md path. Offercheckinstead; regenerate only if the user explicitly asks.adoptable/unwired→ continue.error→ report the failure; do not write anything.
check mode (no writes)
Read the existing DESIGN.md; for every repo-file reference in it, verify the file exists. Report resolvable vs broken pointers, then STOP.
remove mode
Delete the <!-- claudemd-design:begin … <!-- claudemd-design:end --> block from project CLAUDE.md (everything else untouched; show the diff first). Do NOT delete DESIGN.md — report that it stays for the user to remove.
Step 1 — read the real sources
Read every tokenSources[].path from step 0 (first ~200 lines each), plus — when present — whatever injects them (vite css.preprocessorOptions.additionalData, tailwind content, the main.* style import chain). You are extracting FACTS: token names, semantic-color variables, spacing scale, radii, fonts, dark-mode mechanism, reusable mixins / utility classes, contrast annotations.
Step 2 — draft DESIGN.md (thin pointer, facts only)
Contract — every line must trace to something Read in step 1:
- Header: values live in the source files ("以
<file>为准"); DESIGN.md never duplicates full value tables (single source of truth, no drift). - 事实源与加载机制: each token-source path + its load mechanism only if verified (e.g. vite
additionalDataglobal injection ⇒ "$vars/mixins are available in every SFC — do not@usethe file again"). - 硬规则 — evidence-gated menu. Include ONLY rules whose evidence appeared in step 1:
- semantic color variables exist → "use semantic vars, never hardcode hex"
- dark-mode signals → name the mechanism (
html.dark/prefers-color-scheme/darkMode: 'class') + "develop and verify in that theme" - spacing tokens (≥4) → grid rule stating the actual base (e.g. 8px multiples)
- mixin / utility inventory → reuse-first rule listing the real names
- mono font token → "numeric columns use the mono token"
- WCAG / contrast comments → "contrast annotations are commitments; keep them true"
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.
- 2d ago Changed · +2 lines b23cd684729e
- 9d ago First seen · 70 lines · 145 tokens per session scan A 9e0e1b0ff786
claudemd-design-adopt is a command published in the GitHub repository sdsrss/claudemd (2 stars, last pushed today), licensed MIT. It adds 145 tokens to every session and 1,315 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 2026-08-31.
Other commands, from other repositories
interview
Interview your way to an Exolvra Genesis-ready spec and a single-file interactive mockup — or modify an existing pair. Ends with the exact /exolvra-genesis:run line to fire.
verify
Fidelity check — rebuild a page from the extracted tokens, pixel-diff it against the live site, and score how faithfully the tokens capture the design.
simplify
The over-engineering review: five tags (delete, stdlib, native, yagni, shrink), a mandatory replacement per finding, and a real null result when there is nothing to cut.
git
The pre-finish status: branch, hygiene findings, message checks, workflow lint, template state.
init
Install the formatters this repository needs, with every command visible before it runs.
design-tutorial
Interactive guided tour of Naksha — learn commands through real exercises, discover workflows, and get oriented in under 10 minutes.