prototype

A guided command for exploring app ideas by choosing a visual direction, screens, and other prototype details before committing to a roadmap.

In plain words
What is it for?
Use it to select a palette, discuss the product vision, choose prototype screens, and generate an early app concept.
Why use it?
It helps you test and refine an idea through low-risk iterations before spending time building planned features.

Command for Claude Code

Install

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.

agentmods
npx agentmods add commands/marcusgoll/spec-flow/prototype
Clone the repo
git clone --depth 1 https://github.com/marcusgoll/Spec-Flow

Made for: Claude Code.

Per session 19 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 7,813 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00019 $0.07813
Opus 5 $0.00010 $0.03907
Sonnet 5 $0.00004 $0.01563
Haiku 4.5 $0.00002 $0.00781

Measured yesterday against content hash b0699f0d098d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 yesterday.

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/commands/project/prototype.md · 1,025 lines

How it starts

The opening of the file, as written. The whole thing — 1,025 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/prototype - Discovery-First Prototyping (Hybrid Pattern)

v11.0 Architecture: Uses hybrid pattern for discover and explore modes - interactive questions in main context (good UX), prototype generation isolated via Task() (saves ~4-6k tokens).

Related skills (auto-loaded on demand):

  • shadcn-integration - Generates shadcn/ui compatible CSS variables and components.json
  • design-tokens - OKLCH token generation and WCAG validation
  • theme-consistency - Enforces token usage across prototype screens

For discover and explore modes:

User: /prototype discover
         │
         ▼
┌──────────────────────────────────────────────────────┐
│ MAIN CONTEXT (interactive selection - good UX)       │
│                                                      │
│ 1. Check for existing tokens                         │
│ 2. Palette selection (vibe picker)                   │
│ 3. Vision brainstorm question                        │
│ 4. Screen selection (multiselect)                    │
│ 5. Save selections to temp config                    │
│    → .spec-flow/temp/prototype-selections.yaml       │
└──────────────────────────────────────────────────────┘
         │
         ▼
┌──────────────────────────────────────────────────────┐
│ Task(prototype-discover-agent) ← ISOLATED            │
│                                                      │
│ 1. Read selections from temp config                  │
│ 2. Generate theme.yaml from palette                  │
│ 3. Generate theme.css with CSS variables             │
│ 4. Create discovery artifacts (ideas.md, etc)        │
│ 5. Generate HTML for all selected screens            │
│ 6. Generate navigation hub (index.html)              │
│ 7. Update state.yaml                                 │
│ 8. Return summary                                    │
│ 9. EXIT                                              │
└──────────────────────────────────────────────────────┘
         │
         ▼
┌──────────────────────────────────────────────────────┐
│ MAIN CONTEXT (results - minimal tokens)              │
│                                                      │
│ 1. Display success summary                           │
│ 2. Show generated screens                            │
│ 3. Display keyboard shortcuts                        │
│ 4. Suggest next steps                                │
└──────────────────────────────────────────────────────┘

Read the full file on GitHub · 1,025 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. yesterday First seen · 1,025 lines · 0 tokens per session scan A b0699f0d098d

Subscribe to this mod's changes

prototype is a command published in the GitHub repository marcusgoll/Spec-Flow (91 stars, last pushed 4mo ago), licensed MIT. It adds 19 tokens to every session and 7,813 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 2026-08-30.