document

A command that creates a DESIGN.md file describing a project’s visual system, such as colors, fonts, components, and design rules. It builds this description from existing code or from an interview.

In plain words
What is it for?
Use it to document an existing interface or create the project’s visual specification after answering questions about its product and design direction.
Why use it?
It gives later design work a written visual reference so changes remain consistent with the product.

Command

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/adilkalam/orca/document
Clone the repo
git clone --depth 1 https://github.com/adilkalam/orca
Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,571 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 $0.00030 $0.03571
Opus 5 $0.00015 $0.01785
Sonnet 5 $0.00006 $0.00714
Haiku 4.5 $0.00003 $0.00357

Measured yesterday against content hash 71297df11eb5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

document 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.

commands/document.md · 281 lines

How it starts

The opening of the file, as written. The whole thing — 281 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/document — Generate DESIGN.md (Stitch-spec)

Captures the project's visual system as a Stitch-format DESIGN.md so every other design command (/impeccable --craft, /refine, /simplify, /fortify, /motion-design, /recraft) reads the same visual contract.

DESIGN.md is the visual layer (colors, typography, components, do's-and-don'ts). PRODUCT.md is the strategic layer (register, users, anti-references, principles). Run /impeccable --teach first to populate PRODUCT.md; this command requires it as a precondition.

Reference: https://impeccable.style/docs/document and https://stitch.withgoogle.com/docs/design-md/format/

Entry: mandatory skill loading

Before any work, in this order:

  1. Invoke Skill("impeccable-hub") — the aesthetic (felt-state spine + banned rules + preferences + voice-anchors + detector floor). Required for every invocation.
  2. Read {current-project}/.claude/PRODUCT.md. If missing, stop and route to /impeccable --teach first. PRODUCT.md is required input — anti-references and design principles named there are cited in DESIGN.md's Do's-and-Don'ts.

Do not attempt to infer strategic context from the codebase. The user is the only source.

Parse flag

Flag Mode
(none) Scan mode — code already exists; extract tokens from CSS / Tailwind config / design token files; interview only on qualitative language.
--seed Seed mode — no code yet; five-question interview produces a scaffold DESIGN.md with <!-- SEED --> placeholders. Re-run scan once code exists.

If --seed is provided AND {project}/app|src already contains styled code, ask one clarifying question before defaulting (the user might genuinely want a re-scaffold).


Scan mode (default)

Step 0 — Announce what's loaded

Print to user:

PRODUCT.md is loaded. I'll extract visual tokens from the codebase, then ask short qualitative questions to fill the prose sections. The output is a Stitch-spec DESIGN.md (YAML frontmatter + 6 fixed markdown sections in the exact required order).

Read the full file on GitHub · 281 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 · 281 lines · 30 tokens per session scan A 71297df11eb5

Subscribe to this mod's changes

document is a command published in the GitHub repository adilkalam/orca (2 stars, last pushed 1mo ago), licensed MIT. It adds 30 tokens to every session and 3,571 once invoked, about $0.0002 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.