Skip to content

Brooks-Lint

v0.3.0 Breaking

This release includes breaking changes for platform teams planning a safe upgrade.

✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

ai-code-review architecture-review auto-fix claude claude-code claude-code-plugin
+14 more
clean-architecture code-health code-quality code-review code-smells codex-cli-plugin developer-tools gemini-cli-extension llm refactoring skills static-analysis tech-debt test-quality

Summary

AI summary

Adds Documentation Completeness scoring with code-level and arch-level sub-scores.

Full changelog

What's New in v0.3.0

🆕 8th Dimension: Documentation Completeness

Brooks wrote in Ch. 15: "The other face of a program is the documentation it needs." And in Ch. 7: "The Tower of Babel failed not from lack of vision, but lack of communication."

v0.3 adds Documentation Completeness as a full scoring dimension with two independent sub-scores:

  • Code-level — public interfaces, why-comments, parameter clarity
  • Arch-level — ADRs, cross-team interface specs, system README currency

Report format: code-level X/5 · arch-level Y/5 → avg Z/5

🏢 Conway's Law Check (Architecture Audit)

Architecture Audit (Mode 2) now includes a Step 5: Conway's Law Check — an optional section that maps team boundaries against module boundaries and assesses Tower of Babel risk:

  • Structural alignment between team ownership and module structure
  • Cross-team dependency direction (unidirectional vs. bidirectional)
  • Ownerless module detection
  • Common language (interface spec coverage)

📋 ADR Health Assessment (Tech Debt)

Tech Debt Assessment (Mode 3) now classifies ADR debt as a sub-category of Knowledge Debt:

  • Detects missing, stale, or incomplete Architecture Decision Records
  • Repayment guidance: reverse ADRs for untouchable modules, minimum viable ADR template

🌐 Full Internationalization

All skill guide files are now fully in English — no mixed Chinese/English content. The skill is now accessible to international contributors and users.

🔧 12 Quality Fixes

A deep audit of the skill itself identified and fixed 12 issues:

  • Naming consistency: "7 dimensions" corrected to "8 dimensions" throughout
  • Formula clarity: Overall health weighted mean now fully specified with star thresholds
  • Mode detection: Tiebreaker rule for Mode 1 vs Mode 2; Mode 3 covers code-with-debt-question
  • Quick-skip alignment: Dimension 3 scope fixed (codebase-wide, not diff-only); Dimension 5 tightened
  • Auto-generated code: Skip rule added for protobuf stubs, ORM migrations, lock files
  • Architecture guide: Architecture-specific scoring added for all 3 previously missing dimensions
  • Monorepo guidance: How to apply the dependency map and Conway check in monorepos
  • Fallback handling: debt-guide now handles user declining evidence question

Upgrade

# Plugin manager
/plugin install hyhmrright/brooks-lint

# Manual
cp -r skills/brooks-lint ~/.claude/skills/brooks-lint

Breaking Changes

None. The 8th dimension is additive — existing 7-dimension reviews remain valid. Overall health denominator increases from 9 to 10 when all 8 dimensions are scored.

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track Brooks-Lint

Get notified when new releases ship.

Sign up free

About Brooks-Lint

All releases →

Related context

Beta — feedback welcome: [email protected]