Communitygithub.com

TIKAZI/TIKAZ-AI-Skills

TIKAZ AI Skills for Codex — six composable, evidence-first workflow suites

TIKAZ-AI-Skills란 무엇인가요?

TIKAZ-AI-Skills is a Codex agent skill that tIKAZ AI Skills for Codex — six composable, evidence-first workflow suites.

지원 대상~Claude CodeCodex CLI~Cursor
npx skills add TIKAZI/TIKAZ-AI-Skills

즐겨 사용하는 AI에게 물어보기

이 에이전트 스킬이 미리 로드된 새 채팅을 엽니다.

문서

Engineering

Designed, integrated, refactored, and continuously maintained by TIKAZ.

TIKAZ method

  1. Freeze scope, non-goals, acceptance criteria, and evidence required for completion.
  2. Use code-intelligence to map architecture and impact before editing unfamiliar or cross-cutting code.
  3. Use engineering-delivery as the lifecycle owner: plan small slices, implement, test, review, and report.
  4. Route only domain-specific work to supabase-operations or video-workbench.
  5. Run skill-security-audit before importing unfamiliar executable skills or dependencies.
  6. Keep an evidence ledger: commands, results, artifacts, regressions checked, and remaining risk.

Conflict rule

code-intelligence advises; engineering-delivery executes. Domain adapters cannot expand scope without explicit acceptance criteria.

Completion gate

Require repository-native validation, final diff review, no leaked secrets, an explicit release/rollback note when applicable, and objective evidence for every acceptance criterion.

Read references/routing.md and references/output-contract.md.

관련 스킬