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/navraj007in/architecture-cowork-pluginWrote 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/navraj007in/architecture-cowork-plugin/prototype)<a href="https://agentmods.dev/commands/navraj007in/architecture-cowork-plugin/prototype"><img src="https://agentmods.dev/badge/commands/navraj007in/architecture-cowork-plugin/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/navraj007in/architecture-cowork-plugin/prototype"><img src="https://agentmods.dev/badge/commands/navraj007in/architecture-cowork-plugin/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 | $0.00016 | $0.09649 |
| Opus 5 | $0.00008 | $0.04824 |
| Sonnet 5 | $0.00003 | $0.01930 |
| Haiku 4.5 | $0.00002 | $0.00965 |
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 6d 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 — 729 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/architect:prototype
Trigger
/architect:prototype — run after intent.json is created. Works best after blueprint and design-system phases but not required.
Purpose
Generate a fully clickable React prototype that founders can demo to investors, test with users, or validate UX. The prototype must look like a real product — not a developer's wireframe. Distinctive visual design, realistic data, working navigation, and interactive elements.
This is NOT a production application — it's a demo prototype with static/mock data, no backend, but polished visual design and working screen flows.
Output Budget Strategy
A full prototype (10+ screens, 15+ components) exceeds a single output budget. Generation is split into phases using a manifest file. Each phase emits [PROTOTYPE_CONTINUE] — Archon auto-reruns until [PROTOTYPE_DONE].
Workflow
Quick Navigation
| Phase | Steps |
|---|---|
| Setup | Step 0 · Step 1 · Step 1.5 |
| Planning | Step 2 · Step 2.5 · Step 3 |
| Generation | Step 4 · Step 5 · Step 6 · Step 7 |
Step 0: Check Manifest
❓ DECISION POINT: Phase detection (continue or start fresh)
Check if prototype/_manifest.json exists.
If it exists: Read it, determine which phase is complete, and continue from the next phase. Skip to the appropriate step below.
If it does NOT exist: This is Phase 1 — start from Step 1.
Step 1: Gather Inputs
Read (use what's available, don't error if missing):
architecture-output/_state.json— read first if it exists. Use directly:entities→ field names for typed mock datapersonas→ names + roles for realistic UI copy and avatar seedsdesign→ personality, colors, fonts (skip Step 2 derivation if present)mvp_scope.must_have→ which screens/features are in scope
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.
- 6d ago First seen · 729 lines · 16 tokens per session scan A 53ae2aa6715e
prototype is a command published in the GitHub repository navraj007in/architecture-cowork-plugin (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 16 tokens to every session and 9,649 once invoked, about $0.0001 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 2026-09-03.
Other commands, from other repositories
visual
Frontend and UI specialist. Use for React/Vue/Svelte components, CSS/Tailwind styling, accessibility, responsive design, and visual implementation tasks.
design-import
Scaffolds React components from a Claude Design handoff bundle and stops at files on disk: no stories, no tests, no pull request. Use when handed a claude.ai/design URL or a local bundle file; when that same scaffold should carry on through test generation, browser verification and an opened PR, run /ork:design-ship…
healthcare-ux-specialist
Use when designing clinical or patient-facing interfaces. Clinical workflows, displaying sensitive patient data, HIPAA considerations in the UI, and medical terminology patients do not speak.
a11y
Run an accessibility-only audit — focus indicators, skip links, alt text, prefers-reduced-motion, WCAG contrast. Scoped variant of /ui-ux-suite:audit.
web-design
Design and build a complete website or webapp from a project brief — end-to-end workflow from design to working React/Vite code.
checklist
Generate a custom checklist for the current feature based on user requirements.