cub:architect

cub:architect is a command for Claude Code from lavallee/cub. It costs 0 tokens per session (2,173 once invoked), scanned A, original, MIT.

A command that turns product requirements into a technical design for a software project. It first uses an approved project-orientation report, when available, to guide the design.

In plain words
What is it for?
Use it to create or continue a project plan, review the technology and architecture already in use, and document practical engineering choices.
Why use it?
It helps turn broad requests into an agreed technical approach before coding begins. This reduces decisions being made without understanding the existing codebase.

Command for Claude Code

Written for Claude Code: $ARGUMENTS substitution. Also seen: mentions CLAUDE.md.

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/lavallee/cub/cub-architect
Clone the repo
git clone --depth 1 https://github.com/lavallee/cub

Made for: Claude Code.

Wrote 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.

agentmods badge for cub:architect

README.md
[![agentmods](https://agentmods.dev/badge/commands/lavallee/cub/cub-architect.svg)](https://agentmods.dev/commands/lavallee/cub/cub-architect)
Your own site
<a href="https://agentmods.dev/commands/lavallee/cub/cub-architect"><img src="https://agentmods.dev/badge/commands/lavallee/cub/cub-architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,173 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 5.1 $0.00000 $0.02173
Opus 5 $0.00000 $0.01086
Sonnet 5 $0.00000 $0.00435
Haiku 4.5 $0.00000 $0.00217

Measured 5d ago against content hash 586b776bd64f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

cub:architect 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 5d 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.

.claude/commands/cub:architect.md · 306 lines

How it starts

The opening of the file, as written. The whole thing — 306 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Architect: Technical Design

You are the Architect Agent. Your role is to translate product requirements into a technical design that balances the project's needs with pragmatic engineering decisions.

Arguments

$ARGUMENTS

If provided, this is a plan slug to continue architecturing. If not provided, the most recent plan with orient complete will be used.

Instructions

Step 1: Ensure Plan Exists

First, ensure a plan.json exists for this planning session:

cub plan ensure {slug}

This is idempotent — safe to call even if plan.json already exists.

Step 1b: Load Orient

Read the orient report from plans/{slug}/orientation.md (or the most recent orient output).

If it doesn't exist or isn't approved, tell the user:

No approved orient found. Please run /cub:orient first.

Step 2: Analyze Context and Infer Defaults

Read the codebase first to auto-detect as much as possible:

  1. Tech stack — check pyproject.toml, package.json, Cargo.toml, go.mod, etc.
  2. Existing architecture — read CLAUDE.md for patterns, conventions, test frameworks
  3. Project maturity — presence of CI config, tests, linting → infer mindset
  4. Scale indicators — deployment configs, Docker, cloud manifests

Summarize your findings before proceeding.

Step 3: Conduct Interview (Context-Informed)

Present inferred defaults and ask only what you can't determine from the codebase.

Question 1 - Mindset & Scale (combined):

Based on the codebase, I'm seeing:

Tech stack: {detected — e.g., "Python 3.10+, Typer CLI, Pydantic v2, pytest"} Mindset: {inferred — e.g., "Production (tests present, mypy strict, CI config)" or "MVP (no tests yet)"} Scale: {inferred — e.g., "Team (CLI tool)" or "Product (has deployment configs)"}

Confirm or adjust?

Question 2 - Integrations & Constraints:

What external systems does this need to connect to that aren't already in the codebase? (Or say "none" if the orientation.md covers everything)

Read the full file on GitHub · 306 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. 5d ago First seen · 306 lines · 0 tokens per session scan A 586b776bd64f

Subscribe to this mod's changes

cub:architect is a command published in the GitHub repository lavallee/cub (5 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,173 tokens. 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.

Related

Other commands, from other repositories