flutter-review

flutter-review is a skill for Claude Code, Codex from fluttersdk/ai. It costs 234 tokens per session (2,098 once invoked), scanned A, original, MIT.

A review workflow for Flutter code built with the FlutterSDK stack, including the Magic framework and Wind UI. It sends the work to an independent reviewer and then fixes the review findings.

In plain words
What is it for?
Use it to review uncommitted changes, branch differences, or specified Flutter paths, then apply the issues the review identifies.
Why use it?
A second reader can catch assumptions and problems that the original author may overlook.

Skill for Claude CodeCodex

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the fluttersdk plugin — 6 skills, 1 command, 2 agents, 1 MCP server shipped together

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

Made for: Claude Code, Codex.

Or install fluttersdk, the plugin that ships this one along with the rest of its 6 skills, 1 command, 2 agents, 1 MCP server.

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 flutter-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/fluttersdk/ai/flutter-review.svg)](https://agentmods.dev/skills/fluttersdk/ai/flutter-review)
Your own site
<a href="https://agentmods.dev/skills/fluttersdk/ai/flutter-review"><img src="https://agentmods.dev/badge/skills/fluttersdk/ai/flutter-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 234 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,098 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 $0.00234 $0.02098
Opus 5 $0.00117 $0.01049
Sonnet 5 $0.00047 $0.00420
Haiku 4.5 $0.00023 $0.00210

Measured 3d ago against content hash ddadee3ec286, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

flutter-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 3d 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.

skills/flutter-review/SKILL.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Flutter review

A review is worth running when a second reader would catch what the author cannot. That is the whole design here: the reading happens in a subagent with its own context, which never saw the conversation that produced the code, so it cannot inherit the author's assumptions about what the code does.

You orchestrate. You resolve what to review, spawn the reviewer, check its work, and carry the fixes. You do not do the reviewing yourself: reviewing your own recent edits in the same context is the failure this skill exists to avoid.

1. Resolve the scope

Take the first branch that applies:

  1. Arguments name paths or globs (lib/features/monitor/**, two file paths, a feature name that maps to a directory): that is the scope.
  2. git status --porcelain shows uncommitted changes: those files are the scope. This is the common case, and it is what "review my work" means while the work is in progress.
  3. The branch is ahead of its upstream or the default branch: git diff --name-only $(git merge-base HEAD @{u} 2>/dev/null || git merge-base HEAD main)...HEAD is the scope.
  4. Nothing resolves: ask what to review. Do not review the whole lib/ tree on a guess; on a real app that produces a long report nobody reads.

Filter the resolved list to Dart sources. Drop generated files (*.g.dart, *.freezed.dart, _previews.g.dart), and say how many you dropped.

Confirm the project is on this stack before spawning anything: pubspec.yaml names magic or fluttersdk_wind. If it names neither, say so and stop; the rulebooks this review applies would be inventing rules for a project that never adopted them.

Success criterion: a concrete list of Dart files, shown to the user with its origin ("12 uncommitted files", "8 files ahead of main").

2. Pick the reviewer

Reviewer When
fluttersdk:flutter-reviewer The default. A fix, a screen, a handful of files inside one layer.
fluttersdk:flutter-reviewer-deep The change crosses layers (controller plus view plus model, or route plus middleware plus authorization), introduces a pattern the app will copy, or is a whole new feature or module. Also when a standard review came back clean and the user still thinks something is off.

Read the full file on GitHub · 102 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. 3d ago First seen · 102 lines · 234 tokens per session scan A ddadee3ec286

Subscribe to this mod's changes

flutter-review is a skill published in the GitHub repository fluttersdk/ai (2 stars, last pushed yesterday), licensed MIT. It adds 234 tokens to every session and 2,098 once invoked, about $0.0012 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

agents-md-generator

Create, generate, update, repair, trim, or split minimal AGENTS.md files for repository roots and confirmed nested modules using progressive disclosure. Use for missing, bloated, contradictory, or outdated agent instructions, monorepos, services, packages, module context, and portable agent guidance without assuming…

MADTeacher/mad-agents-skills · 69 tokens

flutter-duit-bdui

Integrate, fix, review, migrate, test, or debug Duit / flutterduit backend-driven UI in Flutter applications. Use when a task mentions DUIT, flutterduit, backend-driven UI, BDUI, server-driven UI, XDriver, DuitViewHost, DuitRegistry, HTTP or WebSocket transports, custom Duit widgets, components, capability delegates…

MADTeacher/mad-agents-skills · 94 tokens

flutter-internationalization

Add, fix, audit, and maintain Flutter internationalization with gen-l10n, ARB files, AppLocalizations, flutterlocalizations, intl formatting, plural/select messages, RTL support, locale-specific number/date formatting, and localization build errors. Use when asked to add l10n or i18n, translate Flutter UI text…

MADTeacher/mad-agents-skills · 98 tokens

flutter-animations

Add, fix, refactor, debug, test, or explain Flutter animations and motion effects. Use when working with implicit animations such as AnimatedContainer, AnimatedOpacity, AnimatedSwitcher, and TweenAnimationBuilder; explicit animations using AnimationController, Tween, CurvedAnimation, AnimatedWidget, AnimatedBuilder…

MADTeacher/mad-agents-skills · 109 tokens

flutter-navigation

Implement, fix, refactor, review, migrate, or validate Flutter navigation and routing. Use when working with Navigator, MaterialPageRoute, Router API, gorouter, route guards, redirects, ShellRoute or StatefulShellRoute, nested Navigators, passing and returning route data, deep links, Android App Links, iOS Universal…

MADTeacher/mad-agents-skills · 95 tokens

flutter-adaptive-ui

Build, fix, review, and validate adaptive or responsive Flutter UIs for mobile, tablet, desktop, web, large screens, foldables, and mixed input devices. Use when creating breakpoint-driven layouts, responsive navigation, adaptive dialogs/lists/grids, keyboard/mouse/touch behavior, window-size decisions with MediaQuery…

MADTeacher/mad-agents-skills · 83 tokens