paw-tools-presentation
Generates McKinsey-style HTML presentations from marketing content. Use when user requests to 'create a presentation', 'generate a report', or 'build a deck' from marketing strategy or plan content.
How do I install this agent skill?
npx skills add https://github.com/pawbytes/skill-suites --skill paw-tools-presentationIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The skill is safe. It is designed to generate professional marketing presentations in HTML format using the McKinsey Pyramid Principle. It uses a local Python utility script for file path management and integrates with well-known external services for charting, styling, and stock imagery. The skill includes a human-in-the-loop review of the presentation outline before any files are generated, which serves as a security best practice for processing external content.
- Socketpass
No alerts
- Snykwarn
Risk: MEDIUM · 1 issue
What does this agent skill do?
Presentation Maker
Overview
Transforms marketing strategies, plans, and reports into polished HTML presentations using McKinsey Pyramid Principle / SCR (Situation-Complication-Resolution) structure. Act as a presentation strategist who understands both visual communication and executive-level storytelling.
What it does: Ingests marketing content (from paw-mkt-* skills or raw input), structures it into SCR format, generates a professional HTML document with Chart.js visualizations and hero images.
How it works: Three-stage progressive flow — ingest → structure → generate. Each stage completes before the next begins.
Output: Standalone HTML file with embedded CSS, Chart.js, and a PDF download button at {suite-root}/presentations/. Each presentation has unique visual character — design adapts to content type, audience, and brand signals rather than following a rigid template.
Resolution rules
{skill-root}is this skill's folder; bare paths (references/...,assets/...) resolve from it.{suite-root}isrootfromuv run {skill-root}/scripts/paw-paths.py detect, normally{project-root}/paw-suites/tools. All paths below are relative to it. A custom folder in configtools.presentation_output_folderwins (detectlists it underoverrides; an old default is not one).- When
detectreportsmode: legacyorboth, the user's files are still partly in the old.pawbytes/tools-output/tree. Pass every path throughpaw-paths.py locate <path>(add--writefor a new file) and use the path it returns. Once per session, tell the user thatpaw-tools-setupcan move them topaw-suites/tools/, and offer it. Keep working either way. {brands-folder}is configmkt.brands_folder, default{project-root}/paw-suites/marketing/brands(the marketing suite's brand workspaces;{project-root}/.pawbytes/marketing-suites/brandswhile that suite is not migrated yet).
On Activation
Load available config from {project-root}/.pawbytes/config/config.yaml and {project-root}/.pawbytes/config/config.user.yaml (root keys and the tools section). If config is present, extract:
communication_language(default: English)document_output_language(default: English)
Brand context comes from the marketing suite: {brands-folder}/{brand}/brand-context.md, where {brand} is the brand the user names, else mkt.default_brand, else the only brand folder (ask when there are several). Use it for brand name, audience and voice, and for colors and typography when it states them. An optional {suite-root}/brand-config.json (name, colors, fonts) overrides the visual tokens when present.
Default output path: {suite-root}/presentations/ (or tools.presentation_output_folder from config)
Headless Mode
When invoked with --headless or -H:
- Skip user review of outline
- Use sensible defaults for all choices
- Generate complete HTML output without interaction
- Read brand config if available, otherwise use professional defaults
Stages
Stage 1: Ingest Content
Load ./references/01-ingest.md
Input sources:
- Marketing content from paw-mkt-* skills (passed as
--from-skill <skill-name>) - File path to markdown/text document
- Raw text provided directly
Progression: Content ingested and understood → proceed to Structure
Stage 2: Structure Outline
Load ./references/02-outline.md
Outcome: McKinsey SCR/Pyramid outline presented for user review. User can edit, add, remove, or restructure sections.
Progression: User approves outline → proceed to Generate
Stage 3: Generate HTML
Load ./references/03-generate.md
Outcome: Complete HTML document with:
- Professional styling aligned with brand (if configured)
- Chart.js visualizations for data
- Hero/cover images (Pexels API or Gen AI)
- PDF download button via
window.print()
External Skills Used
- Marketing suite brand workspaces (
brand-context.md) — brand source (optional integration)
Output
Location: {suite-root}/presentations/YYYY-MM-DD-{slugified-title}.html (or under tools.presentation_output_folder).
Filename convention: Today's date plus the slugified title, so files sort by time. Existing presentations keep their names. After writing, run uv run {skill-root}/scripts/paw-paths.py index so the suite README lists it.
Image Sources
Two options available:
- Pexels API — Free stock photos. Requires
pexels_api_keyin config. - fal.ai — AI-generated images. Requires
fal_api_keyin config.
If no API keys configured, use professional placeholder or prompt user to provide images.
Charts
Chart.js is embedded directly in HTML output. Supported types:
- Bar charts (comparisons)
- Line charts (trends over time)
- Pie/doughnut charts (proportions, market share)
- Scatter plots (correlations)
See ./references/chart-patterns.md for implementation patterns.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/pawbytes/skill-suites/paw-tools-presentation">View paw-tools-presentation on skillZs</a>