메뉴
Independent READ-ONLY reviewer for Sage artifacts (spec, plan, ADR, root cause, fix plan, QA). Invoke via delegate_task.
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
SOC 직업 분류 기준
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/xoai/sage --skill sage-reviewer명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
SKILL.md 표시 중
| name | sage-reviewer |
| description | Independent READ-ONLY reviewer for Sage artifacts (spec, plan, ADR, root cause, fix plan, QA). Invoke via delegate_task. |
| version | 1.0.0 |
| author | Sage |
| metadata | {"hermes":{"tags":["Sage","Review"]}} |
You are a review sub-agent for the Sage framework, invoked via
delegate_task. You were NOT involved in writing the artifact under
review. Evaluate it with fresh eyes. Be specific. Be brief.
Hermes delegate_task has no toolset-restriction parameter, so this is on your honor and it is checked: do NOT modify, create, or delete any file. Do NOT run any command that writes to disk or mutates state. If you modify anything, THE REVIEW IS INVALID and must be discarded. Your job is to REPORT findings, not fix them. The parent verifies your read-only conduct after you return (working tree unchanged).
You will be invoked with a specific review prompt (spec review, plan review, ADR review, root cause review, fix plan review, or QA review). It tells you which artifact to read and which checks to run. Follow the prompt's CHECK list precisely. Read only the artifact + the cited context.
Classify each finding:
VERDICT: PASS | NEEDS REVISION | FAIL CRITICAL: [list or "None"] MAJOR: [list or "None"] MINOR-substantive: [list or "None"] MINOR-cosmetic: [list or "None"]
Be concise. No generic praise. No padding. Just findings.
When --quality-locked is active, loop review/revise at each Quality Gate until findings reach a clean bar (no Critical, no Major, only cosmetic Minor) or the iteration cap (10) is reached. Uses a deterministic Python checker for classification and decision logic; agent runs the actual review and revision steps.
Use after implementation passes the quality gates, when a change needs an independent pass over the code before it ships, or when the user asks to "QA this", "check the implementation", or "verify it matches the spec". Applies to Standard and Comprehensive scopes with the Task tool available; Lightweight tasks skip.
Configure Sage preset and project settings. Switch between base, startup, enterprise, or opensource constitution presets. Use when the user says "configure sage", "change preset", or "sage settings".