AI-First SDLC Rules

SkillDev tools

AI-First SDLC compliance rules and standards. Use when checking code quality, validating architecture, enforcing zero technical debt, or making any compliance decision.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the AI-First SDLC Rules skill

What this skill tells your AI

The instructions your AI receives, as published by stevegjones/ai-first-sdlc-practices in skills/rules/SKILL.md and read by ahel’s review.

You are operating under the AI-First SDLC framework. All code must comply with these rules.

For the complete rules reference, see constitution.md.

Quick Reference

Validation Commands

  • After writing code: /sdlc-core:validate --syntax
  • Before commits: /sdlc-core:validate --quick
  • Before PR: /sdlc-core:validate --pre-push

Zero Technical Debt Policy

  • No TODO, FIXME, or HACK comments
  • No any type annotations
  • No commented-out code
  • No deferred fixes or "temporary" solutions
  • Use ./tmp/ not /tmp/

10 Mandatory Logging Points (Application Code)

  1. Function entry/exit with context
  2. Error handling with stack traces
  3. External API/DB calls
  4. State mutations
  5. Security events
  6. Business milestones
  7. Performance anomalies
  8. Configuration changes
  9. Validation failures
  10. Resource limits

Never log: passwords, tokens, PII, biometrics, encryption keys.

Workflow

  1. Feature proposal (docs/feature-proposals/XX-name.md)
  2. Feature branch (feature/name) — never commit to main
  3. Implement with continuous validation
  4. Retrospective (retrospectives/XX-name.md)
  5. PR — CI must pass

Signals

GitHub stars
41
Forks
6
Last commit
Aug 2026
Hacker News mentions
20
Advanced
Catalog kind
skill
Gateway key
rules-stevegjones
Source
github.com/stevegjones/ai-first-sdlc-practices