agentleFS
ww-w-aiww-w-ai/bkit-gemini/skills/plan-plus/SKILL.md

plan-plus

Skill66 starsChanged 7 months agoApache-2.0View on GitHub

Plan Plus — Brainstorming-Enhanced PDCA Planning. Combines intent discovery from brainstorming methodology with PDCA's structured planning. Produces higher-quality Plan documents by exploring user intent, comparing alternatives, and validating incrementally.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

11 rules. 0 are common, 7 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    All output in this skill is documentation and planning artifacts only.

    line 48
  2. Distinctive:

    No source code files may be created or modified until the user explicitly approves the final Plan document.

    line 49
  3. Very specific:

    Read existing documentation in docs/01-plan/, docs/02-design/.

    line 66
  4. Very specific:

    Check .pdca-status.json for current PDCA state.

    line 68
  5. Distinctive:

    Ask the user 4 structured questions with numbered options:

    line 75
  6. Distinctive:

    Ask user to select or combine approaches.

    line 100
  7. Distinctive:

    Remove YAGNI items from the plan scope.

    line 112
  8. Distinctive:

    If the user requests changes to any section, revise and re-present that section.

    line 124
  9. Distinctive:

    After all sections are approved:

    line 128
  10. Very specific:

    Update .pdca-status.json with plan completion.

    line 132
  11. Very specific:

    Create tracker task: [Plan-Plus] {feature}.

    line 133

“Common” is measured: kept by at least 6 in 10 of the 3 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

Plan Plus — Brainstorming-Enhanced PDCA Planning

Produces higher-quality Plan documents by exploring user intent, comparing alternatives, and validating incrementally.

HARD-GATE

Do NOT write implementation code until the Plan is approved by the user.

All output in this skill is documentation and planning artifacts only. No source code files may be created or modified until the user explicitly approves the final Plan document.

Key Principles

  1. Think before coding — Planning prevents wasted effort.
  2. User intent over assumptions — Ask, do not guess.
  3. Minimal viable plan — Cover what matters, skip what does not.
  4. Iterative refinement — Build the plan in stages with user feedback.
  5. Evidence-based decisions — Use research and data, not opinion.
  6. YAGNI discipline — If the user does not need it now, exclude it.

7-Phase Workflow

Phase 0: Project Context Exploration (auto)

Automatically gather project context before interacting with the user.

  1. Read existing documentation in docs/01-plan/, docs/02-design/.
  2. Scan code structure: project root files, src/, lib/, package.json, config files.
  3. Check .pdca-status.json for current PDCA state.
  4. Identify existing features, tech stack, and conventions.

Output: Internal context summary (not shown to user unless requested).

Phase 1: Intent Discovery

Ask the user 4 structured questions with numbered options:

  1. What — What is the feature or change?
    • Provide 3-4 interpretations based on context, let user pick or clarify.
  2. Why — What problem does this solve?
    • Suggest 2-3 possible motivations.
  3. Who — Who are the target users or stakeholders?
    • List likely user personas from project context.
  4. How — What approach or constraints exist?
    • Propose 2-3 technical directions.

Format: Present each question with numbered options. Wait for user response before proceeding.

Phase 2: Alternative Exploration

Based on Phase 1 answers, present 2-3 alternative approaches in a comparison table:

CriteriaApproach AApproach BApproach C
Complexity
Time estimate
Risk level
Scalability
Recommendation

Ask user to select or combine approaches.

Phase 3: YAGNI Review

For each requirement identified so far, classify:

RequirementClassificationRationale
...Must-haveCore to the feature
...Nice-to-haveUseful but not blocking
...YAGNINot needed now, defer

Remove YAGNI items from the plan scope. Move Nice-to-have items to a "Future Considerations" section.

Phase 4: Incremental Design Validation

Present the plan section by section for approval:

  1. Show Objective & Scope — wait for approval.
  2. Show Requirements — wait for approval.
  3. Show Technical Approach — wait for approval.
  4. Show Success Criteria — wait for approval.
  5. Show Risk Assessment — wait for approval.

If the user requests changes to any section, revise and re-present that section.

Phase 5: Plan Document Generation

After all sections are approved:

  1. Generate the full Plan document using templates/plan.template.md.
  2. Save to docs/01-plan/features/{feature}.plan.md.
  3. Update .pdca-status.json with plan completion.
  4. Create tracker task: [Plan-Plus] {feature}.

Phase 6: Next Steps

Guide the user to the next PDCA phase:

  • Recommend /pdca design {feature} to proceed to Design phase.
  • Summarize key decisions made during planning.
  • List any deferred items (Nice-to-have / YAGNI) for future reference.

Arguments

ArgumentDescriptionExample
[feature]Feature name to plan/plan-plus user-auth

References

  • ${extensionPath}/templates/plan.template.md
  • ${extensionPath}/skills/pdca/SKILL.md

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.