This commit replaces the single code reviewer with a structured, three-gate sequential review process to improve quality and efficiency. The new workflow consists of three specialized agents run in order: 1. `code-reviewer` (Gate 1 - Foundation): Validates architecture, code quality, and maintainability. 2. `business-logic-reviewer` (Gate 2 - Correctness): A new agent to verify business rules, requirements, and edge cases. 3. `security-reviewer` (Gate 3 - Safety): A new agent for auditing security vulnerabilities. This sequential process ensures that each stage builds on a validated foundation, preventing wasted effort. For instance, security analysis is only performed on code that has passed architectural and business logic reviews. All reviewer agents are also upgraded to the `opus` model, and all related skills and documentation are updated to reflect this new workflow. --- This change replaces the general-purpose code reviewer with a more robust, three-stage sequential review process involving specialized agents. By introducing distinct gates for code quality, business logic, and security, reviews become more focused and effective. Each gate builds on the previous one, ensuring foundational issues are resolved before more specific analysis begins, which prevents wasted effort and improves final code quality.
6.1 KiB
ring Skills Quick Reference Card
🚨 Most Important Skills (Use These First!)
Before ANY Task
using-ring - Check for relevant skills first (MANDATORY)
Before ANY action → Check skills
Before ANY tool use → Check skills
Before ANY code → Check skills
When Writing Code
test-driven-development - Test first, always
RED → Write failing test → Watch it fail
GREEN → Minimal code → Watch it pass
REFACTOR → Clean up → Stay green
When Something Breaks
systematic-debugging - Find root cause before fixing
Phase 1: Investigate (gather ALL evidence)
Phase 2: Analyze patterns
Phase 3: Test hypothesis (one at a time)
Phase 4: Implement fix (with test)
Before Claiming "Done"
verification-before-completion - Evidence before claims
Run command → Paste output → Then claim
No "should work" → Only "does work" with proof
📚 Available Skills (28 total)
🧪 Testing & Debugging (5)
- test-driven-development - Write test first, watch fail, minimal code
- systematic-debugging - 4-phase root cause investigation
- verification-before-completion - Evidence before claims
- testing-anti-patterns - Common test pitfalls to avoid
- condition-based-waiting - Replace timeouts with conditions
🤝 Collaboration & Planning (9)
- brainstorming - Structured design refinement
- writing-plans - Zero-context implementation plans
- executing-plans - Batch execution with checkpoints
- requesting-code-review - Pre-review checklist
- receiving-code-review - Responding to feedback
- dispatching-parallel-agents - Concurrent workflows
- subagent-driven-development - Fast iteration
- using-git-worktrees - Isolated development
- finishing-a-development-branch - Merge/PR decisions
📋 Pre-Dev Workflow (8 gates)
- pre-dev-prd-creation - Business requirements
- pre-dev-feature-map - Feature relationships
- pre-dev-trd-creation - Technical architecture
- pre-dev-api-design - Component contracts
- pre-dev-data-model - Entity relationships
- pre-dev-dependency-map - Technology selection
- pre-dev-task-breakdown - Work increments
- pre-dev-subtask-creation - Atomic units
🔧 Meta Skills (4)
- using-ring - Mandatory skill discovery
- writing-skills - TDD for documentation
- testing-skills-with-subagents - Skill validation
- sharing-skills - Contributing back
🚀 Commands
- /ring:brainstorm - Interactive design refinement
- /ring:write-plan - Create implementation plan
- /ring:execute-plan - Execute plan in batches
🔍 Review Agents (3 Sequential Gates)
Run in order - each builds on the previous:
-
ring:code-reviewer (Gate 1 - Foundation)
- Architecture, design patterns, code quality, maintainability
- Must pass before Gate 2
-
ring:business-logic-reviewer (Gate 2 - Correctness)
- Domain correctness, business rules, edge cases, requirements
- Must pass before Gate 3
-
ring:security-reviewer (Gate 3 - Safety)
- Vulnerabilities, authentication, input validation, OWASP Top 10
- Final gate before production
Critical: Run sequentially (Code → Business → Security), not in parallel.
📋 By Situation
| Situation | Use This Skill | Key Rule |
|---|---|---|
| Starting new feature | brainstorming |
Research first, ask second |
| Implementing feature | test-driven-development |
Test first, code second |
| Bug appears | systematic-debugging |
Root cause before fix |
| Tests flaky | condition-based-waiting |
No arbitrary timeouts |
| Ready to commit | verification-before-completion |
Evidence before claims |
| Code complete | requesting-code-review |
Gate 1→2→3 sequential |
| Planning work | writing-plans |
Zero-context detail |
| Complex task | pre-dev-* workflow |
8 gates in order |
⚡ Speed Combos
"Build This Feature" Combo
/ring:brainstorm- Design itusing-git-worktrees- Isolate work/ring:write-plan- Plan tasks/ring:execute-plan- Build itrequesting-code-review- Review it (Gate 1→2→3 sequential)
"Fix This Bug" Combo
systematic-debugging- Find root causetest-driven-development- Write test for bugverification-before-completion- Verify fixed
"Major Project" Combo (8 Gates)
pre-dev-prd-creation- Business requirementspre-dev-feature-map- Feature relationshipspre-dev-trd-creation- Technical designpre-dev-api-design- Contractspre-dev-data-model- Data structurespre-dev-dependency-map- Tech choicespre-dev-task-breakdown- Work itemspre-dev-subtask-creation- Atomic tasks
🛑 Universal Rules (All Skills)
State Tracking
SKILL: [name]
PHASE: [current]
COMPLETED: [✓ what's done]
NEXT: [→ what's next]
EVIDENCE: [last output]
When Stuck
- 3 attempts failed? → STOP, reassess
- Confused? → Say "I don't understand X"
- Blocked? → Document blocker explicitly
TodoWrite Required
□ Create todos for each phase
□ Mark in_progress when starting
□ Update after each completion
Exit Criteria
□ All steps complete
□ Verification run
□ Evidence included
□ No "should" or "probably"
🎯 Bulletproofing Features (Nov 2025)
Every skill now enforces:
- Mandatory check points (can't skip)
- Banned phrases (no weasel words)
- Evidence requirements (proof required)
- Time limits (prevent thrashing)
- Failure paths (know when stuck)
🔥 Pro Tips
- Skills are MANDATORY - If one exists for your task, you must use it
- Evidence over claims - Never say "works" without proof
- State tracking - Always know where you are
- Fail fast - 3 attempts max, then escalate
- Zero assumptions - Verify everything
Common Violations to Avoid
❌ "Let me just quickly check..." → Check for skills first ❌ "Should be working now" → Run verification ❌ "I'll test after" → Test FIRST ❌ "Simple fix" → Still need root cause ❌ "Appears correct" → Banned phrase
Remember
Check for skills → Use the skill → Follow it exactly
No exceptions. No shortcuts. No rationalization.