메뉴
Process skill feedback entries — auto-fix obvious prompt issues, bring non-obvious ones to user for discussion.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Scan codebase for dead code, code smells, security issues, contract violations, missing DTOs, and architectural deviations. Creates/updates docs/audit.md and adds actionable items to backlog. Use when user says "audit", "scan", "check code quality", or wants to find hardcoded strings, bypassed abstractions, missing schemas, untyped API boundaries, or drift from shared contracts.
Structured thinking session on a topic. Creates docs/brainstorms/<topic>.md with Status tracking and Action Items. User routes results manually (new sprint, backlog, VISION update, hotfix).
Close the current sprint phase — verify all tasks done, run/write integration tests, advance to next phase.
Close the current sprint — final gate after endgame. Push all commits, update STATUS.md history, update CHANGELOG.
Sprint dispatcher — reads STATUS.md and invokes the right skill based on current sprint state. Use when user says "go", "next", "continue", or wants to proceed with sprint work.
Implement the current sprint task using TDD. Reads task from sprint directory, creates git branch, updates task status on completion. Main development skill.
| name | optimize |
| description | Process skill feedback entries — auto-fix obvious prompt issues, bring non-obvious ones to user for discussion. |
| allowed-tools | Bash, Read, Write, Edit, Grep, Glob |
| argument-hint | [--dry-run] |
Process docs/skill-feedback.md and improve skill prompts based on real execution experience.
--dry-run — show proposed changes without applying themRead docs/skill-feedback.md. Parse all entries below the <!-- entries below --> marker.
Each entry has the format:
## [skill-name] — YYYY-MM-DD
- **Type**: bug | missing-info | optimization
- **Quote**: exact line or section from the skill that caused the issue
- **Problem**: what went wrong or what was missing
- **Suggested fix**: concrete change to the skill text
If there are no entries — STOP: "No feedback to process."
Group entries by skill. For each entry, classify:
Auto-fixable (apply without asking):
make test-all but correct is make test-unit)Needs discussion (bring to user):
Before making any changes, present a summary:
## Optimization Plan
### Auto-fix (will apply)
1. [audit] Fix command: `make test-all` → `make test-unit` (line ~42)
2. [implement] Add missing step: check branch before creating new one
### Needs discussion
3. [plan] "Should /plan auto-run /brainstorm if topic is complex?" — affects scope
4. [triage] "Dedup logic misses renamed tasks" — multiple possible fixes
Wait for user confirmation before applying auto-fixes. Present discussion items one by one.
For each approved auto-fix:
After all entries are processed (applied or discussed), remove them from docs/skill-feedback.md.
Keep the file header and <!-- entries below --> marker.
git add docs/skill-feedback.md .claude/skills/
git commit -m "optimize: process skill feedback — N fixes applied"
## Optimization Report
### Applied
- [skill] change description (entry date)
### Discussed — decided
- [skill] decision made (entry date)
### Discussed — deferred
- [skill] reason for deferral (entry date)
### Discarded
- [skill] reason (entry date)