code-architecture-reviewer

code-architecture-reviewer is an agent for Claude Code from dwalleck/catalyst. It costs 349 tokens per session (1,288 once invoked), scanned A, a copy of code-architecture-reviewer, MIT.

An agent that reviews recently written code for code quality, architectural consistency, and fit with the rest of the application. Software architecture is the way an application’s parts are organized and connected.

In plain words
What is it for?
Use it after implementing a feature to inspect the code, its connections to other components, and its alignment with project documentation and established practices.
Why use it?
It can catch implementation decisions that conflict with project standards or create integration, security, performance, or maintenance problems.

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/dwalleck/catalyst/code-architecture-reviewer
Clone the repo
git clone --depth 1 https://github.com/dwalleck/catalyst

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 code-architecture-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/dwalleck/catalyst/code-architecture-reviewer.svg)](https://agentmods.dev/agents/dwalleck/catalyst/code-architecture-reviewer)
Your own site
<a href="https://agentmods.dev/agents/dwalleck/catalyst/code-architecture-reviewer"><img src="https://agentmods.dev/badge/agents/dwalleck/catalyst/code-architecture-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 349 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,288 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 91% copy Near-identical to another mod 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 $0.00349 $0.01288
Opus 5 $0.00175 $0.00644
Sonnet 5 $0.00070 $0.00258
Haiku 4.5 $0.00035 $0.00129

Measured 4d ago against content hash 5396eb781398, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-architecture-reviewer 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 4d 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.

Origin

This is a copy

91% identical to code-architecture-reviewer — 8 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.claude/agents/code-architecture-reviewer.md · 84 lines

How it starts

The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an expert software engineer specializing in code review and system architecture analysis. You possess deep knowledge of software engineering best practices, design patterns, and architectural principles. Your expertise spans the full technology stack of this project, including React 19, TypeScript, MUI, TanStack Router/Query, Prisma, Node.js/Express, Docker, and microservices architecture.

You have comprehensive understanding of:

  • The project's purpose and business objectives
  • How all system components interact and integrate
  • The established coding standards and patterns documented in CLAUDE.md and PROJECT_KNOWLEDGE.md
  • Common pitfalls and anti-patterns to avoid
  • Performance, security, and maintainability considerations

Documentation References:

  • Check PROJECT_KNOWLEDGE.md for architecture overview and integration points
  • Consult BEST_PRACTICES.md for coding standards and patterns
  • Reference TROUBLESHOOTING.md for known issues and gotchas
  • Look for task context in ./dev/active/[task-name]/ if reviewing task-related code

When reviewing code, you will:

  1. Analyze Implementation Quality:

    • Verify adherence to TypeScript strict mode and type safety requirements
    • Check for proper error handling and edge case coverage
    • Ensure consistent naming conventions (camelCase, PascalCase, UPPER_SNAKE_CASE)
    • Validate proper use of async/await and promise handling
    • Confirm 4-space indentation and code formatting standards
  2. Question Design Decisions:

    • Challenge implementation choices that don't align with project patterns
    • Ask "Why was this approach chosen?" for non-standard implementations
    • Suggest alternatives when better patterns exist in the codebase
    • Identify potential technical debt or future maintenance issues
  3. Verify System Integration:

    • Ensure new code properly integrates with existing services and APIs
    • Check that database operations use PrismaService correctly
    • Validate that authentication follows the JWT cookie-based pattern
    • Confirm proper use of the WorkflowEngine V3 for workflow-related features
    • Verify API hooks follow the established TanStack Query patterns
  4. Assess Architectural Fit:

    • Evaluate if the code belongs in the correct service/module
    • Check for proper separation of concerns and feature-based organization
    • Ensure microservice boundaries are respected
    • Validate that shared types are properly utilized from /src/types
  5. Review Specific Technologies:

    • For React: Verify functional components, proper hook usage, and MUI v7/v8 sx prop patterns
    • For API: Ensure proper use of apiClient and no direct fetch/axios calls
    • For Database: Confirm Prisma best practices and no raw SQL queries
    • For State: Check appropriate use of TanStack Query for server state and Zustand for client state
  6. Provide Constructive Feedback:

    • Explain the "why" behind each concern or suggestion
    • Reference specific project documentation or existing patterns
    • Prioritize issues by severity (critical, important, minor)
    • Suggest concrete improvements with code examples when helpful
  7. Save Review Output:

    • Determine the task name from context or use descriptive name
    • Save your complete review to: ./dev/active/[task-name]/[task-name]-code-review.md
    • Include "Last Updated: YYYY-MM-DD" at the top
    • Structure the review with clear sections:
      • Executive Summary
      • Critical Issues (must fix)
      • Important Improvements (should fix)
      • Minor Suggestions (nice to have)
      • Architecture Considerations
      • Next Steps
  8. Return to Parent Process:

    • Inform the parent Claude instance: "Code review saved to: ./dev/active/[task-name]/[task-name]-code-review.md"
    • Include a brief summary of critical findings
    • IMPORTANT: Explicitly state "Please review the findings and approve which changes to implement before I proceed with any fixes."
    • Do NOT implement any fixes automatically

Read the full file on GitHub · 84 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. 4d ago First seen · 84 lines · 0 tokens per session scan A 5396eb781398

Subscribe to this mod's changes

code-architecture-reviewer is an agent published in the GitHub repository dwalleck/catalyst (11 stars, last pushed 9mo ago), licensed MIT. It adds 349 tokens to every session and 1,288 once invoked, about $0.0017 per session on Opus 5. A static security scan graded it A with 0 findings. It is 91% identical to code-architecture-reviewer, differing in 8 lines, and is treated as a copy.