Pptx

Skill ComeOnOliver/skillshub/skills/happycapy-ai/Happycapy-skills/pptx

🧠 The right skill, one API call. AI agent skills registry with token-efficient skill resolution. 5,000+ skills from 500+ top repos.

Install

npx -y skills add ComeOnOliver/skillshub --skill pptx

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Create polished PowerPoint (.pptx) presentations directly from a topic or content description. Uses HTML as an invisible internal design intermediate — the HTML is never shown to the user. The final and only deliverable is a .pptx file saved to outputs/. Use this skill whenever the user asks to: make a PPT / PowerPoint / 幻灯片 / 演示文稿, generate slides about a topic, or turn any content into a presentation deck. Supports Chinese and English. Always outputs .pptx, never HTML.

SKILL.md

5.8 KB, as published. Nobody here has run it

PPTX Skill

Generate a polished PPTX in three phases. The HTML draft is purely internal — never reference or expose it to the user.

Phase 1 — Design the HTML Draft (Internal only)

Write a single self-contained HTML presentation to tmp/<name>-draft.html.

Design principles:

Choose a bold, distinctive aesthetic (dark editorial, pastel geometry, neon cyber, etc.)
Define a CSS :root {} block with named color variables: --bg, --bg1, --bg2, --text, --muted, --accent, --dim, etc. Use hex values only.
Each <section class="slide" id="sN"> is one slide. Viewport-locked 100vh, scroll-snap, no overflow.
Use Google Fonts — avoid Inter/Roboto. Good choices: Fraunces, DM Sans, Syne, Outfit, Cormorant.
Staggered reveal animations are fine for the HTML but irrelevant to PPTX.

This HTML serves one purpose: capture color theme + slide structure + text content + layout intent before building the PPTX.

Phase 2 — Extract Theme & Build PPTX

Read tmp/<name>-draft.html
Extract the CSS color theme using extract_theme_from_html() from scripts/pptx_builder.py
Write a one-off build script to tmp/<name>-build.py:
- Import Builder from scripts/pptx_builder.py (add skills scripts dir to sys.path)
- Instantiate Builder(theme) with the extracted colors
- Recreate each HTML slide using Builder primitives
- Save to outputs/<name>.pptx
Run: python3 tmp/<name>-build.py

sys.path setup (always required)

import sys
sys.path.insert(0, "/home/node/.claude/skills/pptx/scripts")
from pptx_builder import Builder, extract_theme_from_html
from pptx.util import Inches, Pt
from pptx.enum.text import PP_ALIGN

html = open("tmp/deck-draft.html").read()
theme = extract_theme_from_html(html)   # {"bg":"#0b0b0b", "accent":"#c9a84c", ...}
b = Builder(theme)

Builder API

slide = b.new_slide()              # blank slide with bg fill

# Shapes
b.rect(slide, x, y, w, h, fill="bg1", border="dim")
b.hline(slide, x, y, w, color="accent")
b.oval(slide, x, y, w, h, fill="bg1")
b.card(slide, x, y, w, h)         # shortcut for bg1 card with border

# Text
b.txt(slide, "text", x, y, w, h,
      size=14, bold=False, italic=False,
      color="text",               # theme key or "#hexcode"
      align=PP_ALIGN.LEFT)

b.multicolor_txt(slide, [         # mixed colors in one line
    {"text": "White ",  "size": 52, "bold": True, "color": "text"},
    {"text": "Gold",    "size": 52, "bold": True, "italic": True, "color": "accent"},
], x, y, w, h, align=PP_ALIGN.LEFT)

# Helpers
b.eyebrow(slide, "LABEL", x, y)   # small gold uppercase label
b.logo(slide, "Name", x, y)
b.slide_number(slide, n=1, total=9)

# Grid of cards
b.grid_cards(slide, items, x0, y0, card_w, card_h, gap,
             cols=4, draw_fn=fn)  # fn(slide, b, item, cx, cy, cw, ch)

b.save("outputs/deck.pptx")

Color keys: any key from the theme dict ("bg", "bg1", "text", "muted", "accent", "dim", etc.) or a raw "#rrggbb" string.

Phase 3 — Deliver

Attach outputs/<name>.pptx in the <attachments> block
Do NOT mention the HTML draft or tmp files
The user should only see the PPTX as output

PPTX Layout Conventions

Slide size: 13.33 × 7.5 in (16:9). Set automatically by Builder.

Slide type	Approach
Cover	Oversized multicolor title (Pt 100–130), gold rule, tagline below
Content	Eyebrow + heading + description + card grid
Split	Left column: text + bullets. Right column: visual card panel
Quote	Centered quote Pt 44–60, accent line above
CTA	Centered heading + description + gold button rect + stats row

Limits per slide: 1 heading, max 6 bullets, max 6 cards in a grid.

Common Patterns

Cover with giant title:

s = b.new_slide()
b.multicolor_txt(s, [
    {"text": "My ",  "size": 120, "bold": True, "color": "text"},
    {"text": "Brand", "size": 120, "bold": True, "color": "accent"},
], Inches(0.8), Inches(1.0), Inches(12), Inches(3.5))
b.hline(s, Inches(0.8), Inches(4.8), Inches(2.5))
b.txt(s, "Tagline text here", Inches(0.8), Inches(5.1), Inches(7), Inches(0.9),
      size=16, color="muted")
b.slide_number(s, 1, 9)

Card grid:

items = [("📝", "Title", "Description"), ...]

def draw(slide, b, item, cx, cy, cw, ch):
    ico, lbl, desc = item
    b.txt(slide, ico,  cx+Inches(0.2), cy+Inches(0.2), cw, Inches(0.5), size=20)
    b.txt(slide, lbl,  cx+Inches(0.2), cy+Inches(0.8),  cw-Inches(0.4), Inches(0.45), size=12, bold=True)
    b.txt(slide, desc, cx+Inches(0.2), cy+Inches(1.3),  cw-Inches(0.4), Inches(0.65), size=10, color="muted")

b.grid_cards(s, items, Inches(0.8), Inches(3.4), Inches(2.9), Inches(2.1), Inches(0.22), cols=4, draw_fn=draw)

Bullet list with gold dots:

bullets = ["Item one", "Item two", "Item three"]
for i, text in enumerate(bullets):
    fy = Inches(3.5) + i * Inches(0.55)
    b.oval(s, Inches(0.8), fy + Inches(0.15), Inches(0.12), Inches(0.12), fill="accent")
    b.txt(s, text, Inches(1.1), fy, Inches(5.0), Inches(0.45), size=13)

Gold CTA button:

b.rect(s, Inches(4.7), Inches(5.1), Inches(3.9), Inches(0.6), fill="accent", border=None)
b.txt(s, "Get Started  →", Inches(4.7), Inches(5.13), Inches(3.9), Inches(0.55),
      size=13, bold=True, color="bg", align=PP_ALIGN.CENTER)