무료GitHub에 공개

workflow

Structured work: multi-phase tasks, feature builds, planning, objective loops, hill climbing.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

notque/vexjoy-agent

설치 방법 보기

포함된 파일(133)

SKILL.md12.3 KB
references/adr-template.md9.4 KB
references/agent-upgrade.md12.1 KB
references/article-evaluation-pipeline.md8.9 KB
references/article-evaluation-pipeline/references/report-template.md2.5 KB
references/article-evaluation-pipeline/references/wabi-sabi-classification.md2.4 KB
references/auto-pipeline.md12.4 KB
references/auto-pipeline/references/pipeline-catalog.json15.0 KB
references/auto-pipeline/references/pipeline-catalog.md4.3 KB
references/build_dag.py12.5 KB
references/chain-composer.md19.8 KB
references/chain-composer/references/canonical-chains.md13.9 KB
references/comprehensive-review-workflow.js16.0 KB
references/comprehensive-review.md24.8 KB
references/comprehensive-review/references/output-templates.md10.6 KB
references/comprehensive-review/references/wave-0-per-package.md5.2 KB
references/comprehensive-review/references/wave-1-foundation.md9.3 KB
references/comprehensive-review/references/wave-2-deep-dive.md6.7 KB
references/comprehensive-review/references/wave-3-adversarial.md6.9 KB
references/dag-compatibility-matrix.md13.0 KB
references/dag-composition-patterns.md5.0 KB
references/dag-orchestration-advanced.md6.3 KB
references/dag-orchestration-examples.md11.3 KB
references/dag-skill-patterns.md12.5 KB
references/de-ai-pipeline.js13.9 KB
references/de-ai-pipeline.md6.3 KB
references/discover_skills.py12.1 KB
references/do-perspectives.md11.4 KB
references/doc-pipeline.md9.2 KB
references/domain-research.md31.8 KB
references/domain-research/references/task-type-guide.md11.4 KB
references/explore-pipeline.js17.6 KB
references/explore-pipeline.md15.0 KB
references/fan-out-workflow.js9.3 KB
references/feedback-loop-construction.md14.3 KB
references/fl-design.md9.2 KB
references/fl-error-handling.md8.4 KB
references/fl-implement.md7.4 KB
references/fl-pipeline.md6.0 KB
references/fl-plan.md10.8 KB
references/fl-release.md3.8 KB
references/fl-shared.md3.6 KB
references/fl-validate.md4.5 KB
references/github-profile-rules.md10.3 KB
references/github-profile-rules/references/rule-categories.md6.5 KB
references/hc-domain-playbooks.md15.3 KB
references/hc-ledger.md3.3 KB
references/hc-profiling-tools.md3.7 KB
references/hc-SPEC.md3.3 KB
references/hill-climb.md9.3 KB
references/hook-development-pipeline.md12.9 KB
references/mcp-pipeline-builder.md16.3 KB
references/mcp-pipeline-builder/references/analysis-checklist.md6.2 KB
references/mcp-pipeline-builder/references/design-rules.md7.1 KB
references/mcp-pipeline-builder/references/evaluation-guide.md7.6 KB
references/mcp-pipeline-builder/references/python-scaffold-template.md9.5 KB
references/mcp-pipeline-builder/references/ts-scaffold-template.md1.6 KB
references/mcp-pipeline-builder/scripts/assemble-scaffold.py6.3 KB
references/mcp-pipeline-builder/scripts/register_mcp.py6.9 KB
references/mcp-pipeline-builder/templates/client-cli.ts.tmpl682 B
references/mcp-pipeline-builder/templates/client-http.ts.tmpl1.4 KB
references/mcp-pipeline-builder/templates/index.ts.tmpl1.1 KB
references/mcp-pipeline-builder/templates/package.json.tmpl477 B
references/mcp-pipeline-builder/templates/readme.md.tmpl1.1 KB
references/mcp-pipeline-builder/templates/tools.ts.tmpl3.0 KB
references/mcp-pipeline-builder/templates/ts-scaffold.tmpl556 B
references/mcp-pipeline-builder/templates/tsconfig.json.tmpl408 B
references/ol-SPEC.md2.4 KB
references/ol-state-file.md2.9 KB
references/pipeline-index.json22.2 KB
references/pipeline-retro.md18.7 KB
references/pipeline-scaffolder.md19.2 KB
references/pipeline-scaffolder/references/architecture-rules.md17.3 KB
references/pipeline-scaffolder/references/generated-skill-template.md1.6 KB
references/pipeline-scaffolder/references/pipeline-spec-format.md34.8 KB
references/pipeline-scaffolder/references/step-menu.md21.5 KB
references/pipeline-scaffolder/scripts/assemble-skill.py5.9 KB
references/pipeline-scaffolder/templates/phases/decision-planning.md.tmpl1.2 KB
references/pipeline-scaffolder/templates/phases/domain-extension.md.tmpl1.5 KB
references/pipeline-scaffolder/templates/phases/generation.md.tmpl427 B
references/pipeline-scaffolder/templates/phases/git-release.md.tmpl1.1 KB
references/pipeline-scaffolder/templates/phases/interaction.md.tmpl965 B
references/pipeline-scaffolder/templates/phases/learning-retro.md.tmpl1.3 KB
references/pipeline-scaffolder/templates/phases/observation.md.tmpl1.3 KB
references/pipeline-scaffolder/templates/phases/research-gathering.md.tmpl1.2 KB
references/pipeline-scaffolder/templates/phases/review.md.tmpl930 B
references/pipeline-scaffolder/templates/phases/safety-guarding.md.tmpl1.7 KB
references/pipeline-scaffolder/templates/phases/structuring.md.tmpl361 B
references/pipeline-scaffolder/templates/phases/synthesis-reporting.md.tmpl1.9 KB
references/pipeline-scaffolder/templates/phases/validation.md.tmpl1.1 KB
references/pipeline-scaffolder/templates/skill-template.md.tmpl12.0 KB
references/pipeline-test-runner.md16.5 KB
references/pl-ambiguity-triage.md2.7 KB
references/pl-check-dimensions.md7.4 KB
references/pl-check-errors.md824 B
references/pl-check-verdict-format.md3.2 KB
references/pl-check.md6.5 KB
references/pl-context-boundary.md1.6 KB
references/pl-depth-first-interview.md11.6 KB
references/pl-empirical-prototype.md825 B
references/pl-executor-ready-plan-template.md7.7 KB
references/pl-human-source-elicitation.md1.5 KB
references/pl-manage.md5.1 KB
references/pl-pause-confirm-and-errors.md2.1 KB
references/pl-pause-extract-decisions.md2.9 KB
references/pl-pause-gather-steps.md2.3 KB
references/pl-pause-synthesize-fields.md2.3 KB
references/pl-pause-write-templates.md4.3 KB
references/pl-pause.md6.4 KB
references/pl-plan-files.md8.4 KB
references/pl-pre-plan.md11.6 KB
references/pl-resume.md11.0 KB
references/pl-spec.md5.2 KB
references/pr-pipeline/references/pr-templates.md5.4 KB
references/pr-pipeline/references/preflight-checklist.md2.8 KB
references/pr-pipeline/references/retro-adr-phases.md2.7 KB
references/pr-pipeline/references/review-fix-loop.md1.6 KB
references/research-pipeline.js14.1 KB
references/skill-creation-pipeline.md16.5 KB
references/system-upgrade.md12.4 KB
references/systematic-debugging.md11.0 KB
references/systematic-refactoring.md11.4 KB
references/toolkit-improvement.js28.2 KB
references/toolkit-improvement.md18.2 KB
references/toolkit-improvement/references/adr-template.md10.4 KB
references/toolkit-improvement/references/agent-roster.md15.7 KB
references/workflow-helpers.js4.9 KB
references/workflow-orchestrator.md20.8 KB
references/workflow-orchestrator/references/plan-template.md1.3 KB
references/workflow-orchestrator/references/task-patterns.md23.3 KB
references/workflow-orchestrator/scripts/assemble-plan.py6.9 KB
references/workflow-orchestrator/scripts/validate.py14.4 KB
references/workflow-orchestrator/templates/plan-template.json.tmpl904 B

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Workflow Skill

Five modes for structured multi-phase work. Match the request, follow that mode's instructions.

Mode Selection

Request pattern	Mode
Feature design/plan/implement/validate/release, end-to-end	Feature Lifecycle
Write spec, define requirements, create plan, interview, pause/resume	Planning
Keep working until, iterate until done, drive to done	Objective Loop
Make faster, speed up, reduce latency, profile, hill climb	Hill Climb
All other structured workflows: review, debug, refactor, research, create, explore, upgrade	Ad-Hoc Workflow

Feature Lifecycle

Phase-gated workflow: DESIGN > PLAN > IMPLEMENT > VALIDATE > RELEASE. Each phase must pass its gate before the next begins.

Phase Routing

If .feature/ exists, check state: python3 ~/.claude/scripts/feature-state.py status. Route to the indicated phase.

If no feature state exists, determine entry from intent:

"design", "think through", "explore approaches" -> DESIGN
"plan", "break down", "create tasks" -> PLAN (requires completed design)
"implement", "execute plan" -> IMPLEMENT (requires completed plan)
"validate", "quality gates" -> VALIDATE (requires completed implementation)
"release", "merge", "ship it" -> RELEASE (requires passed validation)
"end to end", "full lifecycle" -> DESIGN (start from beginning)

Phase References

Load the phase reference, then follow it exactly:

Phase	Reference	Produces
DESIGN	`references/fl-design.md`	design.md
PLAN	`references/fl-plan.md`	Wave-ordered task list
IMPLEMENT	`references/fl-implement.md`	Code changes
VALIDATE	`references/fl-validate.md`	Quality gate report
RELEASE	`references/fl-release.md`	Merged PR
End-to-end	`references/fl-pipeline.md`	Full lifecycle
State conventions	`references/fl-shared.md`	--
Error recovery	`references/fl-error-handling.md`	--

State operations use python3 ~/.claude/scripts/feature-state.py only. Never manipulate state files directly.

Planning

Spec writing, plan creation, interviews, ambiguity triage, and session pause/resume. Planning owns specs and saved plans; execution goes through subagent-driven-development or workflow dispatch.

Sub-mode Routing

Signal	Reference
Write spec, user stories, define requirements, scope, acceptance criteria	`references/pl-spec.md`
Discuss ambiguities, resolve gray areas, pre-planning discussion	`references/pl-pre-plan.md`
Interview me, depth-first review, "not sure", "where do I start", "poke holes"	`references/pl-depth-first-interview.md`
Implicit ambiguity or unclear implementation choices	`references/pl-ambiguity-triage.md`
Another person holds needed facts or approval	`references/pl-human-source-elicitation.md`
Observation can settle a disputed choice	`references/pl-empirical-prototype.md`
Phase or session transition near	`references/pl-context-boundary.md`
Create plan, task plan, file-backed planning	`references/pl-plan-files.md`
Check plan, validate plan, pre-execution check	`references/pl-check.md`
List plans, show plan, complete plan, manage plans	`references/pl-manage.md`
Pause, save progress, handoff, stopping for now	`references/pl-pause.md`
Resume, continue, pick up where I left off	`references/pl-resume.md`

For interviews, batch independent questions into frontier rounds. Ask dependent questions sequentially. Include a recommendation per question.

Objective Loop

Iterate-until-verified-done loop. A user states an objective with verifiable done-criteria; each iteration routes one /do cycle, verifies by executing the criteria, and reschedules until verified-done or budget-stop.

Phase 1: SPEC

Gather from the request: objective statement, DONE-CRITERIA (verifiable checks), iteration budget (default 5), NOT-DONE-YET guardrails (what may never be done to satisfy a criterion).

DONE-CRITERIA types: command (preferred -- deterministic command with expected exit code/output) or rubric (only when no mechanical check exists -- frozen at SPEC time, graded by a fresh-context agent).

Phase 2: STATE

Write .objective/<slug>/state.md from references/ol-state-file.md. Wakeups resume from the state file, never conversation memory.

Phase 3: ITERATE

Plan the smallest next step. Route through /do: classify -> route -> dispatch agents -> evaluate. The loop dispatches exclusively through /do -- never edit inline.

Phase 4: VERIFY

Run every done-criterion check. A worker's "passes" claim never substitutes for re-running.

command: run it, paste exit code and output into iteration log.
rubric: dispatch fresh-context sub-agent (did NOT produce the work) with artifact + rubric only. Returns PASS/FAIL with cited evidence.

All pass -> final report, STOP. Any unmet -> Phase 5.

Criteria-gaming guard: a criterion may never be satisfied by weakening a hook, gate, test, or safety control. Stop and report the conflict if that is the only visible path.

Phase 5: RESCHEDULE or STOP

All pass: stop. Unmet + iterations remain: update state file, call ScheduleWakeup (delay 270s for active polling, 1200s+ for idle work). Budget exhausted: honest NOT-DONE report with per-criterion status.

Hill Climb

Metric-driven optimization loop. One number moves; everything else stays fixed. Each iteration: hypothesis -> one change -> correctness floor -> re-measure -> accept or revert.

Phase 1: SPEC

Field	Required	Default
METRIC (one number, units, direction)	yes	--
MEASURE (deterministic command)	yes	--
TARGET (value that ends the loop)	yes	--
FLOOR (correctness gate commands, must exit 0)	yes	--
FIXTURE (pinned dataset/workload)	yes	--
Variance tolerance	no	2x baseline spread
Iteration budget	no	8
Plateau threshold K	no	3

One METRIC per loop. Two numbers with a trade-off: promote one to the FLOOR. Load references/hc-domain-playbooks.md for pre-filled SPEC blocks per domain (frame rate, API latency, CI time, bundle size, memory, token cost).

Phase 2: BASELINE

Run MEASURE N times (N >= 5, N >= 10 for wall-clock). Record median and spread. If spread >= target improvement: STOP -- harness too noisy. Report noise sources and offer to stabilize first.

Phase 3: PROFILE

Locate the cost before changing anything. Load references/hc-profiling-tools.md for per-domain tooling. Guessing at hot spots is the dominant failure mode.

Phase 4: HYPOTHESIZE and CHANGE

State one hypothesis targeting the profiled hot spot. Make one change. Run FLOOR commands -- revert immediately if any fail.

Phase 5: MEASURE

Run MEASURE N times. Compare median to baseline. Accept only if delta > variance tolerance. Update ledger (references/hc-ledger.md). If accepted, new baseline.

Phase 6: LOOP or STOP

Target reached: final report. K consecutive non-improving iterations: plateau stop. Budget exhausted: report what worked and what remains.

Ad-Hoc Workflow

For structured multi-phase work that does not fit the four modes above. Identify the workflow from the table, load its reference, follow its phases exactly.

Cost Gate

Ask first: does this need a multi-agent workflow? Skip the workflow when: single-file mechanical edit (use quick), one agent satisfies the request (direct dispatch), lookup/status/count (direct agent). Escalate only when the request has independent subtasks, needs orthogonal verification, or names "comprehensive / thorough / adversarial / tournament."

Composable Patterns

Pattern	What it does
Classify-and-act	Route by type up front; or classify-at-end
Fan-out-and-synthesize	Independent agents in parallel, barrier, one synthesizer
Adversarial verification	Executor builds, fresh skeptic refutes
Generate-and-filter	Over-generate candidates, gate keeps survivors
Tournament	N agents attempt same task; pairwise judges pick winner per round
Loop-until-done	Repeat until hard completion test passes
Quarantine	Read-only triage agent for untrusted content; separate privileged acting agent

Workflow Catalog

Load the reference for the matched workflow. references/... paths resolve under ${CLAUDE_SKILL_DIR}.

Category	Workflow	Reference
Code Review	Comprehensive multi-wave	`references/comprehensive-review.md`
Debugging	Evidence-based diagnosis	`references/systematic-debugging.md`
Refactoring	Safe refactoring with test gates	`references/systematic-refactoring.md`
Research	Formal research with source gates	`references/research-pipeline.md`
Research	Research to article	`references/research-to-article.md`
Content	Article evaluation	`references/article-evaluation-pipeline.md`
Content	De-AI content	`references/de-ai-pipeline.md`
Content	Documentation	`references/doc-pipeline.md`
Exploration	Codebase exploration	`references/explore-pipeline.md`
Exploration	Multi-perspective analysis	`references/do-perspectives.md`
Creation	Skill creation	`references/skill-creation-pipeline.md`
Creation	Hook development	`references/hook-development-pipeline.md`
Creation	MCP server	`references/mcp-pipeline-builder.md`
Creation	Pipeline scaffolding	`references/pipeline-scaffolder.md`
Creation	Domain research	`references/domain-research.md`
Creation	Chain composition	`references/chain-composer.md`
Creation	Auto-pipeline generation	`references/auto-pipeline.md`
Upgrade	Agent/skill upgrade	`references/agent-upgrade.md`
Upgrade	System upgrade	`references/system-upgrade.md`
Upgrade	Toolkit improvement	`references/toolkit-improvement.md`
Testing	Pipeline test runner	`references/pipeline-test-runner.md`
Testing	Pipeline retro	`references/pipeline-retro.md`
GitHub	Profile rules extraction	`references/github-profile-rules.md`
Orchestration	Task orchestration	`references/workflow-orchestrator.md`
Orchestration	DAG composition	`references/dag-composition-patterns.md`
Orchestration	Compatibility matrix	`references/dag-compatibility-matrix.md`
Orchestration	Common DAG patterns	`references/dag-skill-patterns.md`
Orchestration	DAG examples	`references/dag-orchestration-examples.md`
Orchestration	DAG advanced	`references/dag-orchestration-advanced.md`
Orchestration	Feedback loop	`references/feedback-loop-construction.md`

Terminology

"Workflow" is the canonical term. "Pipeline" is the retained legacy alias -- kept for back-compat in routing keys, meta.name exports, and pipeline-index.json. Use "workflow" in new prose; do not rename code identifiers.

Error Handling

Error	Response
Mode ambiguous	Ask the user to clarify intent
Phase mismatch	Report current state, suggest correct next phase
Missing artifact	Route back to previous phase
Noisy harness (hill-climb)	Stop; report spread vs target improvement
Budget exhausted	Honest NOT-DONE report with per-criterion status
State file missing on wakeup	Report and stop; ask user to restate objective

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

같은 저장소의 스킬

accounts

무료

Manage multiple Claude Code accounts: add, list, check, launch, and install shell aliases for 10+ isolated CLAUDE_CONFIG_DIR profiles.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

notque/vexjoy-agent4422026년 10월 17일 업데이트

architecture-deepening

무료

Improve architecture across modules by deepening interfaces.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

notque/vexjoy-agent4422026년 10월 17일 업데이트

assessment

무료

Assessment: read-only inspection, codebase overview, value analysis, health checks, ADR consultation, decision analysis, multi-perspective critique.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

notque/vexjoy-agent4422026년 10월 17일 업데이트

auto-dream

무료

Background memory consolidation — overnight review, merge, and injection payload for memory files.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

notque/vexjoy-agent4422026년 10월 17일 업데이트

browser-jev-automation

무료

Jev-driven browser automation: Jev picks operations, programs execute, a text model writes field values only when Jev cannot pick one from the goal.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

notque/vexjoy-agent4422026년 10월 17일 업데이트

building-with-jev

무료

Write, compose, integrate, and improve programs that call Jev, TypeSafe's System One judgment model.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

notque/vexjoy-agent4422026년 10월 17일 업데이트

notque의 스킬 모두 보기

이 스킬의 문제 신고

설치

명령으로 설치

npx skills add https://github.com/notque/vexjoy-agent --skill workflow

터미널에 붙여 넣고 실행합니다(Node.js 필요). Claude Code, Codex 등 설치할 에이전트를 고를 수 있습니다.

에이전트에게 부탁하기

다음 Agent Skill을 설치해 주세요.
https://github.com/notque/vexjoy-agent/tree/main/skills/process/workflow

먼저 SKILL.md와 첨부 파일의 내용을 확인해 수상한 처리(외부 전송, 인증 정보 읽기 등)가 없는지 확인한 뒤 Claude Code는 ~/.claude/skills/workflow/, Codex는 ~/.agents/skills/workflow/ 에 배치해 주세요.

이 문장을 그대로 Claude Code나 Codex에 보내면 내용을 확인한 뒤 설치해 줍니다.

GitHub에서 소스 보기

저장소: notque/vexjoy-agent
스타: 442
포크: 50
카테고리: 업무 효율화
라이선스: MIT
원문 언어: 영어
최종 업데이트: 2026년 10월 17일

스킬은 에이전트를 위한 지시와 스크립트의 묶음입니다. cccskills는 내용의 안전성을 보증하지 않습니다. 설치하기 전에 내용을 확인하세요.