design-review

design-review is a skill for Claude Code from ArcBlock/agent-skills. It costs 53 tokens per session (7,568 once invoked), scanned A, original, MIT.

A repeated review of a design or implementation plan by independent reviewers who start each round without the earlier discussion. It produces scores and combines the findings before coding begins.

In plain words
What is it for?
Use it to review a design document, architecture proposal, issue, or directory containing planning files, with a target score and a limit on review rounds.
Why use it?
It can reveal missing requirements, unclear decisions, or risks in a plan before they become implementation problems.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents; names the AskUserQuestion tool; mentions Claude Code.

Part of the agentloop plugin — 20 skills shipped together

Good fit Use it to review a design document, architecture proposal, issue, or directory containing planning files, with a target score and a limit on review rounds.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/arcblock/agent-skills/design-review
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.

Any agent
npx skills add ArcBlock/agent-skills --skill design-review
Clone the repo
git clone --depth 1 https://github.com/ArcBlock/agent-skills

Made for: Claude Code.

Or install agentloop, the plugin that ships this one along with the rest of its 20 skills.

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 design-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/arcblock/agent-skills/design-review/github.svg)](https://agentmods.dev/skills/arcblock/agent-skills/design-review)
Your own site
<a href="https://agentmods.dev/skills/arcblock/agent-skills/design-review"><img src="https://agentmods.dev/badge/skills/arcblock/agent-skills/design-review/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.

agentmods 80×15 button for design-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/arcblock/agent-skills/design-review"><img src="https://agentmods.dev/badge/skills/arcblock/agent-skills/design-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 53 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,568 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Excessive Agency · line 486
    Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
    Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
How audits are shown
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 $0.00053 $0.07568
Opus 5 $0.00026 $0.03784
Sonnet 5 $0.00011 $0.01514
Haiku 4.5 $0.00005 $0.00757

Measured 9d ago against content hash a730711fb9ed, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

design-review 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 9d 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.

plugins/agentloop/skills/design-review/SKILL.md · 504 lines

How it starts

The opening of the file, as written. The whole thing — 504 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Design Review — Iterative Clean-Context Review Loop

Repo profile — read .claude/repo-profile.md first. This skill is repo-agnostic; arc is the reference implementation. Where it references repo identity or paths, read the profile (repo_slug etc.). Arc's own provenance for any lessons is not inlined here (fuller case narratives, where they exist, are under .claude/case-law/).

Automate design document review using clean-context subagents. Each round gets a fresh perspective — no inherited bias from the current conversation.

Usage

/agentloop:design-review <path> [--target <score>] [--max-rounds <n>]
  • <path> — Path to a design document file, or a directory containing design.md + tasks.md
  • --target <score> — Target completeness score (default: 95)
  • --max-rounds <n> — Maximum review iterations (default: 5)

Examples

/agentloop:design-review planning/provider-architecture-rethink/
/agentloop:design-review intent/my-feature/INTENT.md --target 90
/agentloop:design-review planning/my-plan/ --target 95 --max-rounds 3

Issue-driven plans (no planning/ file — issue is the source of truth)

When the plan originated in a GitHub issue (the human confirmed it in the comments), do NOT commit a planning/ doc — those tracker docs rot and become the next audit's deletion. The issue is the source of truth; the file is throwaway:

  1. Render an ephemeral working copy of the confirmed plan (a design.md / tasks.md) into a scratch / gitignored dir (e.g. the session scratchpad) from the issue. Run /agentloop:design-review <scratch-dir> on it. The rounds are disposable iterations on a disposable file.
  2. When it hits the target score, post the final optimized plan back to the issue as a comment — that comment is the durable artifact.
  3. Add a brief round summary: how many rounds, and the key improvement each round made. Then discard the scratch file. Nothing lands in planning/.

Autonomous escalation — ask via an issue comment, never block in-session. In the issue-native flow nobody is babysitting the session, so an escalation that would normally call AskUserQuestion and wait must instead be posted as a comment on the source issue and the run paused there. If the review loop hits an unrecoverable point — a genuine design fork, an AskUserQuestion-worthy ambiguity, a needed architecture change, or an AFS-principle conflict (see the ESCALATION list under "Step 5: Fix Issues") — do not sit waiting for an inline answer. Post a clear, self-contained question comment on the issue (state the options, your recommendation, and what's blocked), then stop that work item. The human answers asynchronously on the issue; the next sweep picks it up. Only fall back to inline AskUserQuestion when a human is demonstrably present and interacting in this session.

Read the full file on GitHub · 504 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. 9d ago First seen · 504 lines · 53 tokens per session scan A a730711fb9ed

Subscribe to this mod's changes

design-review is a skill published in the GitHub repository ArcBlock/agent-skills (5 stars, last pushed yesterday), licensed MIT. It adds 53 tokens to every session and 7,568 once invoked, about $0.0003 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens