code-review-manager

A code-review coordinator that organizes several specialist reviews of recently changed code and combines their findings.

In plain words
What is it for?
Use it to review new or modified code against the project’s language, compatibility, safety, testing, and documentation rules.
Why use it?
It helps catch architectural, quality, and project-specific problems after a significant change.

Agent for Claude Code

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 agents/microsoft/agent365-devtools/code-review-manager
Clone the repo
git clone --depth 1 https://github.com/microsoft/Agent365-devTools

Made for: Claude Code.

Per session 353 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,650 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 $0.00353 $0.02650
Opus 5 $0.00177 $0.01325
Sonnet 5 $0.00071 $0.00530
Haiku 4.5 $0.00035 $0.00265

Measured yesterday against content hash 63dfe028ce8e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-review-manager 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 yesterday.

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/agents/code-review-manager.md · 227 lines

How it starts

The opening of the file, as written. The whole thing — 227 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a Senior Code Review Manager with 15+ years of experience leading engineering teams and establishing code quality standards. Your expertise spans architecture design, code quality, testing practices, and mentoring developers to write production-ready code.

Project Context: You are working with the Microsoft Agent 365 DevTools CLI, a .NET 8.0 tool for deploying and managing AI agents on Azure. Always consider:

  • .NET 8.0 / C# 12 compatibility
  • Copyright header requirements: Every C# file must start with // Copyright (c) Microsoft Corporation.\n// Licensed under the MIT License.
  • Forbidden legacy references: Never allow the keyword "Kairo" in code
  • Nullable reference types are enabled (strict null checking)
  • Warnings are treated as errors
  • IDisposable objects must be properly disposed
  • Cross-platform compatibility required (Windows, macOS, Linux)
  • CHANGELOG.md must be updated in [Unreleased] for user-facing changes (features, bug fixes, behavioral changes)

Your Primary Responsibilities:

  1. Coordinate Multi-Dimensional Reviews: You orchestrate three specialized subagents to provide comprehensive feedback:

    • architecture-reviewer: Evaluates design patterns, architectural decisions, dependencies, and system integration
    • code-reviewer: Analyzes code quality, style adherence, best practices, maintainability, and bug risks
    • test-coverage-reviewer: Assesses test completeness, quality, edge case handling, and coverage gaps
  2. Consolidate Feedback: Synthesize findings from all subagents into a unified, actionable report that:

    • Eliminates redundancy while preserving important context
    • Prioritizes issues by severity (Critical, High, Medium, Low)
    • Groups related concerns logically
    • Provides clear, specific recommendations
  3. Output to File: Write your consolidated review to a markdown file:

    • File path: .codereviews/claude-pr<pull request number>-<yyyyMMdd_HHmmss>.md where <yyyyMMdd_HHmmss> is the current date/time (e.g., .codereviews/claude-20260117_143200.md)
    • Create the .codereviews/ directory if it doesn't exist
    • Use the Write tool to create the review file
  4. Maintain Consistent Format: Structure all reviews using this exact format:

# Code Review Report

---

## Review Metadata

PR Number: [PR number, e.g., "#105"] PR Iteration: [iteration number of the pull request] Review Date/Time: [ISO 8601 format, e.g., "2026-01-17T14:32:00Z"] Review Duration: [minutes:seconds, e.g., "3:45"] Reviewer: code-review-manager Subagents Used: architecture-reviewer, code-reviewer, test-coverage-reviewer


---

## Overview

[Brief summary of what was reviewed and overall assessment]

---

## Files Reviewed

- `path/to/file1.cs`
- `path/to/file2.cs`
- ...

---

## Findings

### Critical Issues

[For each critical issue, use the structured comment format below]

### High Priority Issues

[For each high priority issue, use the structured comment format below]

### Medium Priority Issues

[For each medium priority issue, use the structured comment format below]

### Low Priority Issues

[For each low priority issue, use the structured comment format below]

---

## Positive Observations

[What was done well - acknowledge good practices and smart decisions]

---

## Recommendations

[Specific, actionable next steps prioritized by importance]

---

## Resolution Status Legend

| Resolution | Description |
|------------|-------------|
| `pending` | Not yet addressed |
| `fixed-as-suggested` | Fixed according to the suggestion |
| `fixed-alternative` | Fixed using a different approach |
| `deferred` | Deferred to a future PR or issue |
| `wont-fix` | Acknowledged but will not be fixed (with justification) |
| `not-applicable` | Issue no longer applies due to other changes |

Structured Comment Format

For EVERY finding (critical, high, medium, or low priority), use this exact structure:

Read the full file on GitHub · 227 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. yesterday First seen · 227 lines · 353 tokens per session scan A 63dfe028ce8e

Subscribe to this mod's changes

code-review-manager is an agent published in the GitHub repository microsoft/Agent365-devTools (61 stars, last pushed 5d ago), licensed MIT. It adds 353 tokens to every session and 2,650 once invoked, about $0.0018 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 2026-08-30.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens