MCK PPT Design Skill

AI-native PowerPoint design system — 67 layouts · Harness Engineering · BLOCK_ARC charts · QA pipeline · Python runtime

Copyright © 2024-2026 Kaku Li. Licensed under Apache 2.0. See NOTICE for details.

English · 中文说明 · Harness Guide

---

🖼️ Sample Output

Cover Page	Strategy Analysis	Data Dashboard
4-Column Framework	Color System	Executive Summary

---

⚡ Quick Start

pip install python-pptx lxml

import sys, os
sys.path.insert(0, os.path.expanduser('~/.workbuddy/skills/mck-ppt-design'))
from mck_ppt import MckEngine
from mck_ppt.constants import *

eng = MckEngine(total_slides=12)
eng.cover(title='Q1 2026 Strategy Review', subtitle='Board Presentation', date='2026')
eng.toc(items=[('1', 'Market Overview', 'Current landscape'), ('2', 'Strategy', 'Key actions')])
eng.table_insight(title='Three shifts driving market restructuring',
    headers=['Dimension', 'Before', 'After'],
    rows=[['Distribution', 'Offline-first', 'Digital-first'],
          ['Pricing', 'Cost-plus', 'Value-based']],
    insights=['Digital channels now control 60% of CAC', 'Value pricing unlocks 3× margins'])
eng.donut(title='Revenue Mix', segments=[(0.45, NAVY, 'Product'), (0.35, ACCENT_BLUE, 'Service'), (0.20, ACCENT_GREEN, 'Other')])
eng.timeline(title='12-month roadmap', milestones=[('Q1', 'Foundation'), ('Q2', 'Pilot'), ('Q3', 'Scale'), ('Q4', 'Review')])
eng.closing(title='Thank You')
eng.save('output/deck.pptx')

AI Agent Compatibility

AI Agent	Status
WorkBuddy / Codebuddy	✅ Native skill (mck-ppt-design)
Claude / Claude Code	✅ Load SKILL.md as skill
Cursor / Continue	✅ Add as project rule
Any LLM	✅ Feed SKILL.md as context

---

🔱 Harness Engineering Mode

New in v2.3.3-harness — Transforms the skill from vibe-coding into structured 5-stage generation with machine-readable gates.

The Problem: Vibe Coding vs. Harness

Without Harness, the entire 3,967-line SKILL.md was loaded every time, AI jumped straight to code, and errors were only discovered after rendering. With Harness, context is loaded progressively, structure is locked before content, and gates run as Python scripts — not AI self-evaluation.

Dimension	Without Harness	With Harness
Context load	3,967 lines every time (~6k tokens)	245-line entry → load per-stage (~800 tokens)
Structure design	AI guesses layouts directly	outline.json locks every slide's layout + insight
Action titles	Topic labels ("Market Overview")	Full insight sentences ("Three shifts driving restructuring")
Error discovery	After rendering (expensive to fix)	S3 gate catches API format errors before render
Gate enforcement	AI self-declaration ("looks good")	gate_check.py outputs gate_result.json — passed is a Python bool
Experience accumulation	Discarded after fix	experiences/*.md → AI reads on next deck, avoids repeat mistakes
QA score (A/B test)	95/100 (simple layouts, fewer QA rules triggered)	92/100 (complex layouts, richer content)
Insight titles (A/B test)	0/7 content slides	8/12 content slides