원클릭으로
mb-garden
Clean up and maintain Memory Bank docs so they stay accurate and easy to load.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Clean up and maintain Memory Bank docs so they stay accurate and easy to load.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
Bootstrap a Memory Bank for a new or existing repository, then route into PRD-driven or brownfield workflows.
Optional Analysis before PRD: route vague ideas through facilitated brainstorming and a concise product brief without creating tasks.
Implementation handoff skill for one TASK-* using tier-selected protocol files, local gates, evidence, and explicit verification/sync handoff.
Turn a PRD into a traceable Memory Bank with product, requirements, epics, and features.
Set up deterministic commands, worktrees, and quality gates so agents can run safely in this repository.
Create the Memory Bank skeleton and project command proxies in the current repository.
SOC 직업 분류 기준
| name | mb-garden |
| description | Clean up and maintain Memory Bank docs so they stay accurate and easy to load. |
.memory-bank/ exists and you want to reduce drift, stale docs, or broken routing..memory-bank/.Keep .memory-bank/ accurate, navigable, and cheap-to-prime.
mb-garden owns maintenance and cleanup guidance. The current package also ships deterministic tool assets, but mb-lint and mb-doctor remain separate gates with separate roles.
.memory-bank/ exists.If the repo doesn’t have the tools yet:
scripts/mb-lint.mjs using assets/mb-lint.mjs.scripts/mb-doctor.mjs using assets/mb-doctor.mjs."mb:lint": "node scripts/mb-lint.mjs" and "mb:doctor": "node scripts/mb-doctor.mjs".Current package location includes both tool assets under mb-garden/assets; this is a packaging detail, not conceptual ownership of the doctor readiness role.
Then run:
node scripts/mb-lint.mjs
node scripts/mb-doctor.mjs
Fix all ERROR findings:
tier.memory-bank/tasks/backlog.md or markdown task cards used as workflow staterisk / risk.level task fields; task routing is only by task.tiermb-lint is the structural/mechanical hygiene gate. mb-doctor is the workflow/autonomous readiness gate over mb-lint. Default mb-doctor is suitable for normal health checks and fresh skeletons. Run strict mode only after the JSON task queue exists: after /prd-to-tasks, before scheduler execution inside /autonomous, or before /autopilot when the queue is already prepared:
node scripts/mb-lint.mjs
node scripts/mb-doctor.mjs --strict
Do not treat strict mode as required for a bare generated skeleton with an empty task registry.
architecture WHAT/WHY + guides HOW) or by equivalent spec-driven support docs..memory-bank/archive/.If the repo uses GitHub Actions:
.github/workflows/memory-bank-lint.yml from assets/memory-bank-lint.yml.Write a summary (what changed + what remains) to:
.tasks/TASK-MB-GARDEN/TASK-MB-GARDEN-S-01-final-report-docs-01.mdRun this checklist weekly (or every 5–10 meaningful changes):
.memory-bank/**/*.md has description + status.architecture/<X>.md + guides/<X>.md pair or by clearly routed equivalent spec-driven support docs.requirements.md RTM matches actual feature/test status..task.json, every task has tier, and no orphaned tasks exist without a feature link..memory-bank/changelog.md has entries for recent changes.index.md and router-indexes resolve correctly.archive/ has a tombstone stub at original location.node scripts/mb-lint.mjs passes (0 errors).node scripts/mb-doctor.mjs readiness gate passes in default mode; --strict passes post-queue before scheduler/autopilot execution.