agentleFS
ww-w-aiww-w-ai/bkit-codex/.agents/skills/pdca/SKILL.md

pdca

Skill69 starsChanged 8 months agoApache-2.0View on GitHub

Unified PDCA cycle management. Supports: plan, design, do, analyze, iterate, report, status, next, archive, cleanup. Actions: plan->create plan doc, design->create design doc, do->implementation guide, analyze->gap analysis (design vs code), iterate->auto-fix gaps, report->completion report, status->show progress, next->suggest next phase. Triggers: pdca, plan, design, analyze, iterate, report, status, gap analysis, 계획, 설계, 분석, 검증, 보고서, 計画, 設計, 分析, 计划, 设计, planificar, diseño, analizar, planifier, conception, Planung, pianificare Do NOT use for: bkit rules reference (use $bkit-rules), template details (use $bkit-templates), level-specific guidance (use $starter/$dynamic/$enterprise).

We split this skill into the rules it gives an agent

21 rules. 0 are common, 12 are distinctive, and 9 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 12

    Distinctive

    This author's own choices. Worth a look.

  • 9

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create a planning document for the feature.

    line 64
  2. Very specific:

    Call bkit_pdca_plan(feature, level) MCP tool

    line 68
  3. Very specific:

    Write template to docs/01-plan/features/{feature}.plan.md

    line 69
  4. Very specific:

    Call bkit_complete_phase(feature, "plan") when done

    line 71
  5. Distinctive:

    Create a technical design document based on the plan.

    line 92
  6. Distinctive:

    Verify plan exists (required prerequisite)

    line 100
  7. Distinctive:

    Verify design exists (required prerequisite)

    line 130
  8. Distinctive:

    Follow implementation order from design

    line 132
  9. Very specific:

    Call bkit_pre_write_check(filePath) before each file write

    line 133
  10. Very specific:

    Call bkit_post_write(filePath, linesChanged) after significant changes

    line 134
  11. Very specific:

    Read design document: docs/02-design/features/{feature}.design.md

    line 159
  12. Very specific:

    Write analysis to docs/03-analysis/{feature}.analysis.md

    line 163
  13. Very specific:

    If matchRate >= 90%: suggest $pdca report {feature}

    line 167
  14. Distinctive:

    Read gap list from analysis document

    line 191
  15. Distinctive:

    Stop conditions: matchRate >= 90% OR maxIterations reached

    line 200
  16. Distinctive:

    If max iterations reached without 90%: report current state and suggest user review

    line 201
  17. Distinctive:

    Verify matchRate >= 90% (warn if below)

    line 215
  18. Distinctive:

    Verify report completion (phase = "completed" or matchRate >= 90%)

    line 290
  19. Very specific:

    Update .pdca-status.json: phase = "archived"

    line 293
  20. Distinctive:

    Read archived features from status

    line 310
  21. Distinctive:

    Delete selected features from status

    line 312

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

PDCA Unified Skill

Unified Skill for managing the complete PDCA cycle: Plan, Design, Do, Check, Act.

Usage

$pdca plan {feature}      Create plan document
$pdca design {feature}    Create design document
$pdca do {feature}        Implementation guide
$pdca analyze {feature}   Gap analysis (Check phase)
$pdca iterate {feature}   Auto-improvement (Act phase)
$pdca report {feature}    Completion report
$pdca status              Current PDCA status
$pdca next                Next phase suggestion
$pdca archive {feature}   Archive completed PDCA
$pdca cleanup             Clean archived features

Phase Flow

Plan -> Design -> Do -> Check(analyze) -> Act(iterate) -> Report -> Archive
  |                                          |
  |                                          v
  |                                    (if < 90%)
  |                                    iterate -> re-analyze
  |                                          |
  v                                          v
[Complete]                             (if >= 90%)
                                       Report -> Archive

Phase Progress Visualization

[Plan] -> [Design] -> [Do] -> [Check] -> [Act] -> [Report]

Status indicators:
  [Phase] done     -> phase completed
  [Phase] active   -> currently working
  [Phase] pending  -> not yet started

Plan Phase

Create a planning document for the feature.

Steps

  1. Call bkit_pdca_plan(feature, level) MCP tool
  2. Write template to docs/01-plan/features/{feature}.plan.md
  3. Fill in sections: Goals, Scope, Success Criteria, Schedule
  4. Call bkit_complete_phase(feature, "plan") when done

Key Sections

  • Overview and Purpose
  • Scope (In/Out)
  • Functional Requirements
  • Non-Functional Requirements
  • Success Criteria (Definition of Done)
  • Risks and Mitigation
  • Architecture Considerations
  • Convention Prerequisites

Output Path

docs/01-plan/features/{feature}.plan.md


Design Phase

Create a technical design document based on the plan.

Prerequisites

Plan document must exist. If missing, suggest: $pdca plan {feature} first.

Steps

  1. Verify plan exists (required prerequisite)
  2. Call bkit_pdca_design(feature, level) MCP tool
  3. Write template to docs/02-design/features/{feature}.design.md
  4. Fill in: Architecture, Data Model, API Spec, Test Plan
  5. Call bkit_complete_phase(feature, "design") when done

Level-Specific Templates

LevelTemplateFocus
Starterdesign-starter.template.mdSimple structure, learning-oriented
Dynamicdesign.template.mdFull-stack with BaaS integration
Enterprisedesign-enterprise.template.mdMSA, Clean Architecture, K8s

Output Path

docs/02-design/features/{feature}.design.md


Do Phase

Guide implementation based on the design document.

Prerequisites

Design document must exist. If missing, suggest: $pdca design {feature} first.

Steps

  1. Verify design exists (required prerequisite)
  2. Reference design document during implementation
  3. Follow implementation order from design
  4. Call bkit_pre_write_check(filePath) before each file write
  5. Call bkit_post_write(filePath, linesChanged) after significant changes
  6. Call bkit_complete_phase(feature, "do") when done

Implementation Guide

Provide a structured implementation plan:

  1. Data Layer: Types, models, API client
  2. Business Logic: Services, custom hooks, state management
  3. UI Components: Base components, pages, error handling
  4. Integration: Connect API to UI, loading states, error handling

Output Path

docs/02-design/features/{feature}.do.md (optional implementation tracking)


Analyze Phase (Check)

Compare design document vs implementation to find gaps.

Steps

  1. Call bkit_pdca_analyze(feature) MCP tool
  2. Read design document: docs/02-design/features/{feature}.design.md
  3. Scan implementation code in relevant directories
  4. Compare design items vs implemented items
  5. Calculate Match Rate: (implemented / total_design_items) * 100
  6. Write analysis to docs/03-analysis/{feature}.analysis.md

Decision After Analysis

  • If matchRate >= 90%: suggest $pdca report {feature}
  • If matchRate < 90%: suggest $pdca iterate {feature}

Gap Categories

CategoryDescriptionExample
MatchDesign and code alignAPI endpoint exists as designed
Missing in CodeDesigned but not implementedMissing validation logic
Missing in DesignImplemented but not designedExtra utility function
ChangedImplemented differentlyDifferent data structure

Output Path

docs/03-analysis/{feature}.analysis.md


Iterate Phase (Act)

Auto-fix identified gaps to improve match rate.

Steps

  1. Read gap list from analysis document
  2. Fix identified gaps in code (prioritize "Missing in Code" items)
  3. Re-run analysis: $pdca analyze {feature}
  4. Repeat until matchRate >= 90% or max 5 iterations
  5. Call bkit_complete_phase(feature, "act") when done

Iteration Rules

  • Maximum iterations: 5 (configurable via bkit.config.json)
  • Stop conditions: matchRate >= 90% OR maxIterations reached
  • If max iterations reached without 90%: report current state and suggest user review

Report Phase

Generate a completion report for the feature.

Prerequisites

Match rate should be >= 90%. Warn if below.

Steps

  1. Verify matchRate >= 90% (warn if below)
  2. Gather data from plan, design, analysis documents
  3. Generate completion report
  4. Write to docs/04-report/{feature}.report.md
  5. Include: completed items, quality metrics, learnings

Report Sections

  • Summary (completion rate, duration)
  • Related Documents (plan, design, analysis links)
  • Completed Items (functional, non-functional requirements)
  • Quality Metrics (match rate, code quality score)
  • Lessons Learned (keep, problem, try)
  • Next Steps

Output Path

docs/04-report/{feature}.report.md


Status

Show current PDCA progress for all active features.

Steps

  1. Call bkit_get_status(feature) MCP tool
  2. Display feature, phase, matchRate, iteration count
  3. Show progress visualization

Output Format

PDCA Status
---
Feature: {feature}
Phase: {current_phase}
Match Rate: {rate}%
Iteration: {count}/{max}
---
[Plan] done -> [Design] done -> [Do] done -> [Check] active -> [Act] pending

Next

Suggest the next PDCA phase based on current state.

Steps

  1. Call bkit_pdca_next(feature) MCP tool
  2. Display recommended next action with command

Phase Guide

CurrentNextCommand
Noneplan$pdca plan {feature}
plandesign$pdca design {feature}
designdoStart implementation
docheck$pdca analyze {feature}
check (<90%)act$pdca iterate {feature}
check (>=90%)report$pdca report {feature}
reportarchive$pdca archive {feature}

Archive

Archive completed PDCA documents.

Steps

  1. Verify report completion (phase = "completed" or matchRate >= 90%)
  2. Create docs/archive/YYYY-MM/{feature}/ folder
  3. Move documents from original locations
  4. Update .pdca-status.json: phase = "archived"

Documents Archived

  • docs/01-plan/features/{feature}.plan.md
  • docs/02-design/features/{feature}.design.md
  • docs/03-analysis/{feature}.analysis.md
  • docs/04-report/{feature}.report.md

Cleanup

Clean up archived features from .pdca-status.json.

Steps

  1. Read archived features from status
  2. Display list with timestamps
  3. Delete selected features from status
  4. Archive documents remain in docs/archive/

Template References

Templates are available in references/ directory:

TemplatePurpose
plan.template.mdPlan document structure
design.template.mdDesign document structure
design-starter.template.mdStarter-level design
design-enterprise.template.mdEnterprise-level design
analysis.template.mdGap analysis structure
report.template.mdCompletion report structure
do.template.mdImplementation guide structure

Download now (ZIP)

  • Other versions7 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 7

    1. ww-w-ai/bkit-claude-code★ 605same name
    2. popup-studio-ai/bkit-claude-code★ 601same name
    3. twhsi/skills★ 259same name
    4. ww-w-ai/bkit-gemini★ 66same name
    5. unclejobs-ai/second-claude-code★ 41same name
    6. AlexeyShirshov/skills★ 15same name
    7. tupe12334/instinct★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • Used alongsideProjects with it also use pm-discovery, qa-phase and 7 more.

    Other files people use alongside this one.

    Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
    • ww-w-aiSkill

      pm-discovery

      PM Agent Team - Automated product discovery, strategy, and PRD generation. Runs PM analysis workflow to produce comprehensive PRD before PDCA Plan phase.

      Also in 3 projects that use this

    • ww-w-aiSkill

      qa-phase

      Run QA tests at various levels (L1-L5) for features under PDCA management. Integrates with MCP tools to execute structured test plans.

      Also in 3 projects that use this

    • kannaka-labsSkill

      skill-status

      Show where you are in the workflow and what to do next — use for progress checks and orientation. Use when: AUTOMATICALLY ACTIVATE when user asks about:. \"status\" or \"progress\" or \"where am I\". \"what's next\" or \"next step\"

      Also in 3 projects that use this

    • WelluxSkill

      rollback

      Execute or plan a deployment rollback. Invoke for: "rollback", "revert deployment", "undo deploy", "something broke in prod", "roll back to previous version", "deployment failed", "revert to v{version}".

      Also in 3 projects that use this

    • dacryptSkill

      control

      Control your Tesla — lock, unlock, climate, trunk, horn, flash, windows, sentry mode, charge port, speed limit. Use when someone says "lock my car", "turn on the AC", "open the trunk", "honk the horn", etc.

      Also in 3 projects that use this

    • launch52-aiSkill

      skill-create

      Create and validate Claude Code skills with proper structure, checklists, and validation scripts. Use when creating a new skill, auditing existing skills, or ensuring skills meet quality standards. Generates complete skill scaffolds with SKILL.md, check scripts, guides, and templates.

      Also in 3 projects that use this

    • +1 more from ww-w-ai
  • In this repository27 other files it gives its agents.

Does it work?

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

Reviews can't be read right now.

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.