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/w2ur/claude-code-setupWrote 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/w2ur/claude-code-setup/new-app)<a href="https://agentmods.dev/commands/w2ur/claude-code-setup/new-app"><img src="https://agentmods.dev/badge/commands/w2ur/claude-code-setup/new-app/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/w2ur/claude-code-setup/new-app"><img src="https://agentmods.dev/badge/commands/w2ur/claude-code-setup/new-app.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 | $0.01812 |
| Opus 5.5 | $0.00006 | $0.00725 |
| Sonnet 5.5 | $0.00003 | $0.00362 |
| Haiku 4.5 | $0.00002 | $0.00181 |
Grade A, and why
new-app 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 — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create a new portfolio app named $0 in ~/Dev/$0.
Validation
- Name format:
$0must be kebab-case. If it contains uppercase, spaces, or underscores, reject and ask for a corrected name. - Name collision: check if
~/Dev/$0already exists. If yes, abort. - GitHub collision: run
gh repo view {github-username}/$0 2>/dev/null— if the repo already exists, warn.
Information Gathering
Before scaffolding, ask the owner:
- What does this app do? (one paragraph)
- Who is it for? (public / family / personal)
- Preferred stack? (or "you choose")
- User-facing language? (French / English / Bilingual)
- Deploy target? (Netlify / Vercel / Cloudflare / local-only for now)
- Should it get a tile on the hub ({portfolio-site-url})? If yes, get a one-line French tagline, one-line English tagline, and (if it's a real project the hub would feature, not a personal-only tool) a facts line for each language.
- Design direction? (or I propose 2-3) If the owner has one, take it as given. Otherwise load the
design-directionskill and run its steps 1-5 on the answers above (question 1 is the subject), and wait for the owner to pick one before scaffolding. Its step 6 (DESIGN.md) runs at scaffold time, below; its step 7 runs once there is a page to render. An app with no UI (CLI, script, library) may answer "none"; record that instead.
Scaffold
Using the answers above, create:
Repository setup
mkdir -p ~/Dev/$0
cd ~/Dev/$0
git init
DESIGN.md
Unless question 7 was answered "none", write DESIGN.md at the repo root as the design-direction skill's step 6 says (from its references/design-md-template.md). If the owner gave a direction, fill the template from it. The CLAUDE.md "Design direction" section below is the pointer line that step 6 asks for.
CLAUDE.md
Create covering: project overview, tech stack, dev/build commands, deployment, and any conventions that override or extend the global CLAUDE.md. Include:
- Project overview from the owner's description
- Tech stack from chosen stack
- User-facing language
- Development commands (npm install && npm run dev, or equivalent)
- Deployment info
- A Design direction section from question 7 — later UI sessions read it instead of falling back on a model default. Write the chosen direction concretely, then the avoid-list and the trigger line; do not copy the skill's full list. The avoid-list is not written in this command: read
~/.claude/CLAUDE.md→ "Design defaults" at scaffold time and copy its named list of patterns (the items after "Never default to") verbatim, so the new project holds its own self-contained copy of the current list:
If question 7 was answered "none", write the section as a single line:## Design direction [Chosen direction in one or two sentences: palette, type, the one signature element.] The full direction is in `DESIGN.md`, this project's identity. Avoid unless this section asks for it: [the list, copied verbatim from ~/.claude/CLAUDE.md → Design defaults]. **Before any UI or visual work here, load the `design-tells` skill** — it holds the full trap list. This section overrides it wherever they disagree.No UI yet — before adding one, load the design-direction skill.
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 Changed · +3 lines 163c98035546
- 21d ago Changed · +11 lines 611d3451163c
- 1mo ago First seen · 129 lines · 16 tokens per session scan A 3a0fed2771d1
new-app is a command published in the GitHub repository w2ur/claude-code-setup (2 stars, last pushed 3d ago), licensed MIT. It adds 16 tokens to every session and 1,812 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 2026-08-31.
Other commands, from other repositories
execute
Execute the next available tracked task with TDD, pre-commit validation, PR creation, AI code review, and issue tracker bridge sync.
phase
Orchestrate a full roadmap phase. Designs features, imports to tracker, creates issues, then executes.
dashboard
Compact visual progress dashboard with progress bar, task table, dependency graph, and active/blocked/next status.
plan-start
5-phase planning command: PRD analysis, design review, technical decisions, dynamic research team, metrics. Produces a complete implementation plan + ADRs before any code is written.
status
Show current execution status with progress, dependencies, file ownership, and parallelization opportunities.
sync
Reconcile coco tracker state with the configured issue tracker. Syncs task statuses and reports mismatches.