Optimize your web pages so they get cited by ChatGPT, Gemini, and Claude
Write and restructure pages so ChatGPT, Gemini, and Claude quote them — not a competitor’s.
A free Agent Skill that drafts, critiques, and upgrades slide decks with consulting-grade structure — action titles, the Pyramid Principle, and Situation-Complication-Resolution. Try it in Claude or ChatGPT in one click.
Free to try. One click loads the skill into Claude or ChatGPT, with no account and no setup.
A McKinsey-style deck is a presentation built as an argument, not a slideshow. The name comes from the methodology developed at McKinsey & Company and now standard across top-tier consulting firms — but the ideas are older than any firm and belong to anyone who has to convince a room.
The defining rule is the action title. On an ordinary slide, the headline names a topic — "Revenue Overview," "Market Landscape." On a consulting slide, the headline states the takeaway as a complete sentence. Stack those sentences and the titles alone — read top to bottom, ignoring every chart and bullet — tell the entire story. An executive skimming your deck in the elevator gets the argument in thirty seconds. That is the whole trick, and it is why these decks win decisions.
It works because it respects how busy people actually read: conclusion first. A consulting deck leads with the recommendation, then supports it (the Pyramid Principle), and carries the audience along a Situation → Complication → Resolution arc. A normal deck buries the point on slide 14; a McKinsey deck puts it on slide 2 and spends the rest earning it. The difference isn’t polish — it’s that one is designed to be understood and the other is designed to be presented.
Revenue Overview
Revenue declined 15% year-over-year, driven by three operational failures
Market Landscape
The addressable market has doubled since 2022, but three incumbents control 74% of spend
Any time a deck has to change a decision, not just inform. If the audience can walk out having learned something but done nothing, structure barely matters. The moment money, headcount, or strategy hinges on the room saying yes, the argument has to be airtight — and that’s where this methodology pays for itself.
Most slide decks decorate. A consulting deck argues. This skill applies the methodology pioneered at McKinsey and adopted across top-tier firms so your deck reads as an argument an executive can follow from the titles alone.
It works in three modes. Critique — hand it an existing deck and it evaluates the narrative, the titles, and each slide’s logic, and tells you exactly what to fix. Draft — describe the situation and it builds a storyline from scratch. Update — point it at specific problems and it applies the fixes. It’s a Productivity-category skill and it’s free.

Real prompts, and what the skill hands back.
“Critique this 20-slide board deck. Read only the titles in sequence — do they tell a coherent story? Flag every topic title and rewrite it as an assertion.”
A slide-by-slide critique against narrative, title, and logic dimensions — with every “Q3 Results”-style topic title rewritten into a takeaway sentence.
“Draft a ghost deck for a Series A pitch: situation, complication, resolution, then three proof pillars. Action titles only, no design yet.”
A storyboard of full-sentence action titles that tells the whole argument top-to-bottom before a single slide is designed.
Every headline is a complete sentence stating the takeaway. Read the titles top to bottom and you have the whole argument.
Lead with the answer, then group support into MECE pillars — each level proves the one above it (Barbara Minto’s method).
The deck’s arc: shared context, the tension that demands action, then your recommendation.
Each slide makes exactly one point — the one in its title. Every chart, table, and icon exists to prove it.
No slide leaves the audience thinking “so what?” — the implication is always explicit.
Lay out the full sequence of action titles before any design. If the titles don’t tell a story, the deck isn’t ready.
Slide Craft gets the argument right; a style skill gets the look right. Stack it with a deck-styling skill — the way our own team pairs it with agentman-pptx-style — and you go from a rough storyline to an on-brand, exhibit-driven deck in one pass.
It also pairs naturally with research and analysis skills: let a synthesis skill find the story, then let Slide Craft structure it into an argument an executive will actually approve.
The methodology is a consulting firm’s core product, and this skill puts it in your own hands for free. Here’s how to run it on a deck you already have (or one you’re starting from scratch):
Already connected your library to Claude? Tell it: “Use the mckinsey-slide-craft skill from the agentman_skills MCP server.”
Not connected yet? The Try in Claude / ChatGPT button loads it from the public library, with no setup. Or connect your library.
Click “Try in Claude / ChatGPT” above, or — if you’ve connected your library — tell Claude: “Use the mckinsey-slide-craft skill from the agentman_skills MCP server.” Naming the slug and server points it at exactly the right skill.
Tell it whether you want to Critique an existing deck, Draft a new one, or Update one with specific fixes. If you don’t say, it’ll ask.
Paste your slide titles (or describe your situation) and ask it to read them in sequence: do they tell a coherent story? It will flag every topic title and rewrite it as an action-title sentence.
Before any design, get the storyline right: title → executive summary → situation → complication → resolution → next steps. The skill lays this out as action titles first, so you approve the argument before investing in slides.
For each slide, ask it which chart fits — it applies a chart-selection guide (comparison → bar, change over time → line, part-of-whole → waterfall, never pie) and flags anti-patterns like data dumps and bullet-point disease.
Ask it to run the “so what?” sweep: every slide has to make its implication explicit. If a skeptical executive would think “so what?”, the skill tells you where and how to fix it.
That’s the same sequence a consultant would run — the difference is it costs you a prompt instead of a retainer, and you keep the methodology for the next deck.
One click loads the skill into Claude or ChatGPT. No account, no setup.
Three things above all: action titles (every headline is a full-sentence takeaway, so the titles alone tell the story), the Pyramid Principle (lead with the answer, then MECE supporting pillars), and a Situation-Complication-Resolution arc. This skill applies all three, plus the “one message per slide” and “so what?” tests.
Yes — that’s its Critique mode. Paste or describe your existing deck and it evaluates the narrative structure, the titles, and each slide’s logic, then tells you specifically what to change. It also has Draft mode (build from scratch) and Update mode (apply targeted fixes).
Its focus is the argument — titles, storyline, and what each slide should prove. For the visual layer, stack it with a deck-styling skill (our team pairs it with agentman-pptx-style) so structure and design come together on brand.
No. Click “Try in Claude” or “Try in ChatGPT” above and the skill loads from our public content API — no account or setup. Clone it into an Agentman workspace to customize it and share it with your team.