brooks-lint: a complementary skill for deep architectural review (pairs well with caveman)

[hyhmrright]

Hey caveman community! 👋

I've been using caveman to keep token usage lean while working through large codebases — great tool. I wanted to share something I built that pairs nicely with it.

brooks-lint is a Claude Code skill that reviews code through the lens of 12 classic software engineering books — Fowler, Martin, Brooks, Evans, Ousterhout, and more. While caveman optimizes how you communicate with Claude, brooks-lint focuses on what architectural issues to look for.

They complement each other well in practice:

• Use caveman to keep Claude's context lean and efficient
• • Use brooks-lint to get architectural feedback grounded in proven principles
    Example output on a PaymentService:

[Clean Code] processPayment() does 4 things — violates SRP
[Release It!] No timeout on payment gateway — cascade failure risk  
[DDIA] No idempotency key — retries will double-charge
[Pragmatic Programmer] Validation logic duplicated in 3 places — DRY violation

GitHub: https://github.com/hyhmrright/brooks-lint

Would love to hear if anyone tries combining the two — caveman for token efficiency + brooks-lint for design review.

[hyhmrright]

Just published a more detailed write-up on Dev.to that covers all three tools together — caveman + brooks-lint + logic-lens as a complete stack:

https://dev.to/hyhmrright/caveman-brooks-lint-logic-lens-the-complete-ai-code-review-stack-for-claude-1e3e

Includes a full comparison table showing how each tool covers a different layer (efficiency vs architecture vs runtime correctness), plus real output examples from all three.