matteocervelli

planning-doc-generator

16
1
# Install this skill:
npx skills add matteocervelli/llms --skill "planning-doc-generator"

Install specific skill from multi-skill repository

# Description

Generate project assessment markdown documents from JSON data with WHY/WHO/WHAT

# SKILL.md


name: planning-doc-generator
description: Generate project assessment markdown documents from JSON data with WHY/WHO/WHAT
sections and GO/NO-GO decision matrix.
allowed-tools: Read, Write, Edit


Planning Document Generator Skill

Purpose

Generate structured assessment documents from JSON configuration. Converts project planning data into markdown assessment reports with purpose, stakeholder, and scope analysis plus a GO/NO-GO decision framework.

When to Use

  • Creating project assessment documents
  • Generating planning documentation from structured data
  • Building evaluation reports with decision matrices
  • Documenting project vision and scope
  • Creating stakeholder alignment assessments
  • Generating baseline project documentation

Input: JSON Format

The skill expects JSON input with the following structure:

{
  "project_name": "Project Name",
  "date": "2025-11-03",
  "why": {
    "exists": "Why does this project exist?",
    "problem": "What problem does it solve?",
    "vision": "What is the desired outcome?"
  },
  "who": {
    "stakeholders": "List of key stakeholders",
    "decision_makers": "Who decides",
    "executors": "Who does the work",
    "concerns": "Their priorities and concerns"
  },
  "what": {
    "building": "What are we building/changing?",
    "features": "Key features and components",
    "out_of_scope": "What is out of scope",
    "success_criteria": "Definition of success"
  },
  "go_no_go": {
    "purpose_clarity": "✓|⚠|✗",
    "stakeholder_alignment": "✓|⚠|✗",
    "scope_definition": "✓|⚠|✗",
    "resource_availability": "✓|⚠|✗",
    "timeline_feasibility": "✓|⚠|✗",
    "risk_assessment": "✓|⚠|✗",
    "success_metrics": "✓|⚠|✗"
  },
  "decision": "GO|CONDITIONAL|NO-GO",
  "rationale": "Explanation of decision"
}

Template Filling Process

  1. Load templates/assessment-template.md
  2. Replace all {PLACEHOLDER} values with JSON data
  3. Calculate coverage: Count non-empty answers ÷ 17 questions
  4. Insert status indicators (✓/⚠/✗) from GO/NO-GO section
  5. Generate markdown with formatted decision matrix
  6. Validate all sections populated with content (no {ANSWER} remaining)

Coverage Calculation

Total question count: 17

Breakdown:
- WHY section: 3 questions
- WHO section: 4 questions
- WHAT section: 4 questions
- GO/NO-GO section: 7 assessment items
- Other: 1 additional (missing info summary)

Formula:

Coverage = (Number of answered/populated fields ÷ 17) × 100
Percentage = Round to nearest whole number

Output Location

Generated documents save to:

~/docs/planning/{project_slug}/assessment-{date}.md

Example:

~/docs/planning/project-name/assessment-2025-11-03.md

Workflow

JSON Input
    ↓
Load Template
    ↓
Replace Placeholders
    ↓
Calculate Coverage
    ↓
Format Decision Matrix
    ↓
Validate Completeness
    ↓
Write to ~/docs/planning/
    ↓
Markdown Output

Key Features

Status Indicators

  • Green: Ready to proceed
  • Yellow: Proceed with caution / clarification needed
  • Red: Blocker / do not proceed

Decision Framework

  • GO: All indicators green, proceed immediately
  • CONDITIONAL GO: Some yellow flags, proceed with mitigation
  • NO-GO: Red flags present, do not proceed without resolution

Coverage Tracking

Automatically calculates and displays:
- Number of questions answered (X/17)
- Percentage coverage
- List of missing information

Best Practices

  1. Complete All Fields: Aim for 100% coverage (17/17)
  2. Be Specific: Use concrete details, not generic placeholders
  3. Stakeholder Buy-in: Ensure WHO section reflects actual decision-makers
  4. Realistic Assessment: Be honest in GO/NO-GO evaluation
  5. Document Decisions: Clear rationale essential for tracking

Example Usage

# Command-line usage
planning-doc-generator \
  --input project-plan.json \
  --output ~/docs/planning/myproject/

# Result
~/docs/planning/myproject/assessment-2025-11-03.md

Integration Points

Input Sources

  • Project planning worksheets (converted to JSON)
  • Kickoff meeting notes (structured into JSON)
  • Requirements documents (parsed to JSON format)
  • Stakeholder surveys (aggregated to JSON)

Output Consumers

  • Project stakeholders (for review/approval)
  • Project managers (for tracking)
  • Decision makers (for GO/NO-GO calls)
  • Documentation archives (for reference)

Validation Rules

Before writing output file:
- All {PLACEHOLDER} values replaced
- No {ANSWER} tokens remaining
- Project name populated
- Date populated (YYYY-MM-DD format)
- Decision field contains valid value (GO, CONDITIONAL, NO-GO)
- Coverage calculated and accurate


Version: 1.0.0
Created: 2025-11-03
Scope: Global utility skill

# Supported AI Coding Agents

This skill is compatible with the SKILL.md standard and works with all major AI coding agents:

Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.