Adelie-SquadAdelie-Squad/solosquad/skills/workflow-refinement/SKILL.md

workflow-refinement

Skill25 starsChanged 8 days agoMITView on GitHub

Workflow 종료 시 개선 제안. stage 의존성 / handoff 효율 / 평균 cycle time 분석. Chief 가 RETROSPECT 에서 호출.

In this repository32 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- build-testSkill
- trend-trackerSkill
- typecheckSkill
- agent-managerSkill
- cron-managerSkill
- design-systemSkill
- discovery-synthesisSkill
- docsSkill
- experiment-designSkill
- goal-managerSkill
- hypothesis-designSkill
- interview-scriptSkill
- jobs-storiesSkill
- lean-canvasSkill
- market-researchSkill
- meceSkill
- okrSkill
- opportunity-treeSkill
- policySkill
- prdSkill
- premortemSkill
- primitive-reviewSkill
- prioritizationSkill
- retrospectiveSkill
- skill-managerSkill
- skill-refinementSkill
- triageSkill
- wbsSkill
- workflow-managerSkill
- xyz-hypothesisSkill

The file

Create a free account to mount

Files

SKILL.md

Workflow Refinement Skill

Workflow 종료 시 어떤 stage 가 bottleneck 인지, 어떤 handoff 가 정보 손실 큰지 식별.

입력

종료된 workflow id
<org>/workflows/<id>/_status.yaml
<org>/workflows/<id>/stage-*/handoff_*.md
<org>/memory/ledger/<task-id>.jsonl

분석 차원

per_stage:
  - stage_id: "..."
    duration_min: <N>
    spawns: <N>
    open_questions: <N>
    handoff_artifact_size_lines: <N>
    next_stage_blocked_min: <N>   # 다음 stage 시작까지 대기 시간

bottlenecks:
  - "stage-3-design (75% of total cycle time)"

handoff_quality:
  - from_stage: "stage-2-plan"
    to_stage: "stage-3-design"
    fidelity: high | medium | low   # 다음 stage 가 추가 질문 없이 시작했는지
    issues: ["..."]

출력

workflow_id: "..."
cycle_time_total: <min>
bottleneck_stages: [...]
handoff_issues: [...]
refinement_proposals:
  - target: stage-3-design
    proposal: "researcher → researcher 사이에 explicit handoff template 도입"
    expected_impact: "다음 stage 평균 대기 -40%"
    effort: small

≥ 2 approaches (gstack)

각 proposal 에 대안 1개 이상.

Anti-Sycophancy

❌ "workflow 가 매끄럽게 진행되었습니다"
✅ "stage-3 평균 4시간 대기. 원인 = handoff 누락. fix A vs B 비교."

Reference

v1.1 PRD §5.2 RETROSPECT

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free