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.
git clone --depth 1 https://github.com/mahmoudilyan/marmouiWrote 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/mahmoudilyan/marmoui/prototype)<a href="https://agentmods.dev/commands/mahmoudilyan/marmoui/prototype"><img src="https://agentmods.dev/badge/commands/mahmoudilyan/marmoui/prototype/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/mahmoudilyan/marmoui/prototype"><img src="https://agentmods.dev/badge/commands/mahmoudilyan/marmoui/prototype.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 | $0.00036 | $0.00916 |
| Opus 5.5 | $0.00014 | $0.00366 |
| Sonnet 5.5 | $0.00007 | $0.00183 |
| Haiku 4.5 | $0.00004 | $0.00092 |
Grade A, and why
prototype 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 2mo 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/marmo-ui:prototype
Scaffold a new prototype using @marmoui/ui, wire up the
@marmoui/prototype-overlay (Anthropic OAuth sign-in + inline comments
- owner-only select-mode prompt-to-clipboard), and deploy to Netlify.
Run every step. Don't skip the MCP validation pass — prototypes are read by reviewers from outside the engineering team and any hallucinated prop or broken composition is going to surface in a comment thread inside an hour.
Steps
-
Confirm prerequisites.
node --version≥ 20pnpm --version≥ 9netlify --version(suggestnpm i -g netlify-cliif missing)NPM_TOKENset (required to install@marmoui/*from GH Packages)MARMO_PROTOTYPE_OAUTH_CLIENT_IDandMARMO_PROTOTYPE_COMMENTS_APIset in shell env (provisioned once per developer, point at the team Supabase project and Anthropic OAuth client). If missing, print the one-time setup instructions fromapps/prototype-starter/README.mdand stop.
-
Choose a target directory outside the monorepo. Default:
~/Sites/marmo-prototype-<kebab-case-name>. If the dir exists and is non-empty, ask for a different name. -
Copy the starter template. Source =
apps/prototype-starter/in themarmo-uimonorepo (resolve viamcp__marmo-ui__get_knowledge_version().repoRootif available, otherwise ask the user for the path). Copy excludingnode_modules,dist,.turbo. Rewrite thepackage.jsonnamefield to match the new directory. -
Wire env vars. Create
.env.localfrom.env.examplewith:VITE_COMMENTS_API=$MARMO_PROTOTYPE_COMMENTS_API VITE_ANTHROPIC_OAUTH_CLIENT_ID=$MARMO_PROTOTYPE_OAUTH_CLIENT_ID -
Generate the screen from the user's
$ARGUMENTSdescription.- Call
mcp__marmo-ui__get_knowledge_versiononce. - Call
mcp__marmo-ui__get_design_guidelinesonce. - Pick the closest pattern via
mcp__marmo-ui__get_patternif one matches; otherwisemcp__marmo-ui__search_componentsandmcp__marmo-ui__get_component_infoper component. - Write the JSX into
src/App.tsx, replacing the placeholder. - Imports only from
@marmoui/ui+react-icons/md. - Run
mcp__marmo-ui__review_generated_codeon the final file. Fix every error and warning untilvalid: true. Never ship code with errors.
- Call
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.
- 2mo ago First seen · 83 lines · 36 tokens per session scan A 7f5fb12fd194
prototype is a command published in the GitHub repository mahmoudilyan/marmoui (19 stars, last pushed 3mo ago), licensed MIT. It adds 36 tokens to every session and 916 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
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.
design-tutorial
Interactive guided tour of Naksha — learn commands through real exercises, discover workflows, and get oriented in under 10 minutes.
ux-audit
Audit a Figma file against a design brief — checks page structure, frame naming, sizes, styles, components, and content compliance.
gen-video
Generate video output in one of three modes: Remotion (programmatic React-based video), Gen-AI (shot prompt pack for Runway/Kling/Sora/Pika/Luma), or Remotion+Gen-AI (programmatic scaffold with AI footage embedded).
critique-ux
Run a focused UX critique on a screen — affordances, information density, and hierarchy — and output a prioritised fix list.
design-onboarding
Design a first-run experience end to end — activation path, progressive disclosure, and time to first value.