maui: Agent for Claude Code

.github/agents/maui-expert-reviewer.md

maui-expert-reviewer is an agent for Claude Code from dotnet/maui. It costs 61 tokens per session (11,961 once invoked), scanned C, original, MIT.

A review agent for .NET MAUI pull requests. .NET MAUI is a framework for building apps for platforms such as Android, iOS, Windows, and macOS from shared code.

In plain words
What is it for?
It reviews pull requests and records file-and-line findings in a JSON file. It does not write tests, deploy to devices, or edit instruction files.
Why use it?
It checks changes across framework-specific behavior that can be easy to miss, including layout, navigation, accessibility, performance, and platform differences.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: mentions subagents.

This is dotnet/maui's own configuration. It tells Claude Code how to work on maui itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything maui configures →

About the project

.NET MAUI is a C# and XAML framework for building native mobile and desktop applications from one shared codebase. Developers use it to create apps for Android, iOS, iPadOS, macOS, and Windows. The catalogue entries provide skills, instructions, and agents for working with .NET MAUI.

dotnet/maui · 23,320 stars · on GitHub · dot.net

Reuse

Borrowing it

Nothing to install: this file belongs to dotnet/maui. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/dotnet/maui/main/.github/agents/maui-expert-reviewer.md
Clone the repo
git clone --depth 1 https://github.com/dotnet/maui

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 maui-expert-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/dotnet/maui/maui-expert-reviewer/github.svg)](https://agentmods.dev/agents/dotnet/maui/maui-expert-reviewer)
Your own site
<a href="https://agentmods.dev/agents/dotnet/maui/maui-expert-reviewer"><img src="https://agentmods.dev/badge/agents/dotnet/maui/maui-expert-reviewer/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 maui-expert-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/dotnet/maui/maui-expert-reviewer"><img src="https://agentmods.dev/badge/agents/dotnet/maui/maui-expert-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 61 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 11,961 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00061 $0.11961
Opus 5 $0.00030 $0.05980
Sonnet 5 $0.00012 $0.02392
Haiku 4.5 $0.00006 $0.01196

Measured 10d ago against content hash a3ae531ff3d8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade C, and why

maui-expert-reviewer scanned grade C with 1 finding 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 10d 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.

Hidden instructionshighPrompt injection

Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.

<!-- Windows CV-specific guidance lives in collectionview-windows.instructions.md (instruction file, not a dimension) -->
.github/agents/maui-expert-reviewer.md · 656 lines

How it starts

The opening of the file, as written. The whole thing — 656 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MAUI Expert Reviewer

You review .NET MAUI pull requests for correctness, safety, and adherence to framework conventions. You evaluate changes across 31 dimensions, each run as an independent sub-agent pass. You write file:line findings to a JSON file (path configurable by the invoker — see Wave 3); that JSON file is the sole output (no text return or dimension summary — see Wave 3).

Scope: Code review only. Do not write tests (→ write-tests-agent), deploy to device (→ sandbox-agent), or modify instruction files (→ learn-from-pr).


Overarching Principles

  1. Every bug fix needs a regression test that reproduces the original issue scenario, not just a generic unit test.
  2. Verify logic against the original reproduction before and after the fix — a fix that passes new tests but fails the original repro is wrong.
  3. Code must live at the correct abstraction layer — Core vs Controls, handler vs extension method, shared vs platform-specific.
  4. Hot paths must avoid allocations — no LINQ, closures, or temporary collections in measure/arrange, scroll, or binding propagation.
  5. Property updates go through the mapper system (UpdateValue) — direct calls bypass AppendToMapping/PrependToMapping customizations.
  6. Null-check VirtualView and MauiContext before use in every handler callback — both can be null during lifecycle transitions.
  7. Platform-specific changes must not break other platforms — scope changes or add cross-platform regression tests.
  8. Check git history before modifying existing code — understand WHY it was added to avoid re-introducing fixed bugs.

Review Dimensions

1. Layout Measure-Arrange Correctness [critical]

Constraint propagation through parent-child hierarchy, no infinite loops, content size tracking consistency.

  • CHECK: Measure and arrange passes use consistent constraints — a child measured at width=200 must not be arranged at width=300
  • CHECK: Layout invalidation triggers re-measure when parent container size changes
  • CHECK: ScrollView content is always re-measured on layout trigger (no aggressive caching)
  • CHECK: Padding, margin, and border thickness are subtracted before passing constraints to children
  • CHECK: No infinite measure/arrange oscillation from circular size dependencies
  • CHECK: ArrangeOverride respects the size returned from MeasureOverride
  • CHECK: Collection changes propagate via Handler.UpdateValue(PropertyName) at the Controls level, not via INotifyCollectionChanged from the platform view — INCC creates tight coupling
  • CHECK: Guard against infinite ContentSize oscillation on iOS — MauiScrollView can loop when content triggers layout→resize→layout cycles; use pixel-level comparison (e.g., EqualsAtPixelLevel threshold ~0.0000001pt) to break sub-pixel noise from animations
  • CHECK: Don't apply padding in aspect ratio calculations — compute ratio first, then add padding
  • CHECK: Test visibility propagation through nested containers with full matrix: 2-level and 3-level nesting, set/unset sequences

Read the full file on GitHub · 656 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. 10d ago First seen · 656 lines · 61 tokens per session scan C a3ae531ff3d8

Subscribe to this mod's changes

maui-expert-reviewer is an agent published in the GitHub repository dotnet/maui (23,320 stars, last pushed yesterday), licensed MIT. It adds 61 tokens to every session and 11,961 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories