agentleFS
jsuchetjsuchet/jonathan-skills-lab/skills/consulting-deck/SKILL.md

consulting-deck

Skill1 starsChanged 5 months agoMITView on GitHub

Build consulting-grade presentations following McKinsey/BCG best practices with MECE storylines, Minto pyramid structure, and professional slide writing. Use this skill whenever the user asks to create, structure, outline, or write a presentation, deck, pitch, storyline, lecture, training, or slides, whether from scratch or refining existing content. Also trigger when the user mentions MECE, Minto pyramid, slide writing, consulting deck, storyline structure, deck outline, presentation brief, slide copy, slide structure, or asks for help at any stage of presentation creation (briefing, storyline, outline, individual slides, or review). This skill handles the full workflow from strategic brief through final slide content, and can be entered at any stage. Works for any brand or audience: brand skills (e.g. your brand skill) can be layered on separately.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. About 0 are common, 11 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If unsure, default to /consulting-deck.

    line 15
  2. Distinctive:

    If unsure, ask: "Do you want me to challenge and refine what you have, or build this from the ground up?"

    line 47
  3. Very specific:

    For .pptx: Use the pptx skill (read pptxgenjs.md or editing.md depending on whether a template exists).

    line 53
  4. Distinctive:

    Default to English. If the user requests French, produce all slide copy in French.

    line 59
  5. Distinctive:

    MECE: Every grouping at every level must be Mutually Exclusive and Collectively Exhaustive.

    line 65
  6. Distinctive:

    Minto Pyramid: Start with the answer. Group supporting arguments.

    line 66
  7. Distinctive:

    Slide titles are messages: Every slide title is a complete sentence stating the slide's key takeaway.

    line 67
  8. Distinctive:

    So-what test: Every piece of data, chart, or statement must pass the "so what?" test: the audience should immediately understand why it …

    line 69
  9. Distinctive:

    No em dashes anywhere. Use colons, semicolons, or line breaks.

    line 76
  10. Distinctive:

    No hype words: revolutionary, leverage, synergy, game-changing, cutting-edge, transform (when vague), optimize (when vague).

    line 77
  11. Distinctive:

    When completing a stage, always:

    line 82
  12. Distinctive:

    Ask if the user wants to continue to the next stage or stop here

    line 86
  13. Distinctive:

    Never auto-advance between stages.

    line 88

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Consulting Deck Skill

Build consulting-grade presentations using proven frameworks: MECE storylines, Minto pyramid logic, McKinsey/BCG-level slide writing. Layered with [YOUR NAME]'s own training and presentation pedagogy (orientation devices, audience-keyed structuring, theory-exercise rhythm, hook-via-question openings).

When to Use This vs. /slide-deck

  • Use /consulting-deck (this skill) for: client-facing strategic decks, training and workshop content, lectures, recommendations, pitches, board reviews. Anything that carries reputational weight or will be presented externally. Enforces full MECE, Minto, and message-style title discipline.
  • Use /slide-deck for: quick rough outlines for operational decks, internal status updates, fast proposals, throwaway slides. When speed matters more than rigour.

If unsure, default to /consulting-deck.

How This Skill Works

This skill covers five stages. The user can run the full workflow or jump into any stage independently. Detect which stage the user needs based on what they provide:

StageNameUser signalsReference file
1Brief"new deck", "presentation for...", "I need a deck about..."references/stage-1-brief.md
2Storyline"help with storyline", "structure my narrative", has a brief or objectivereferences/stage-2-storyline.md
3Outline"outline", "slide breakdown", "detailed structure", has a storylinereferences/stage-3-outline.md
4Slides"write the slides", "slide content", "copy for slide X", has an outlinereferences/stage-4-slides.md
5Review"review my deck", "check the flow", "quality check", has a draft deckreferences/stage-5-review.md

Always read the relevant reference file(s) before starting work on a stage. Multiple stages can be read if the user wants to run a full or partial workflow.

Also always read references/consulting-principles.md: it contains the foundational MECE and Minto frameworks that apply across all stages.

For slide layout recommendations, read references/slide-type-library.md during Stages 3-4.

For [YOUR NAME]'s recurring presentation devices (hook openings, orientation map slides, audience-keyed structuring, theory-exercise rhythm, etc.), read references/author-pedagogy.md during Stages 2-4. These layer on top of the universal consulting principles, not in place of them.

For deck-type-specific starter templates (lecture, training/workshop, recommendation, pitch, status update, kick-off), read references/deck-types-cookbook.md during Stages 1-2 to jump faster to a sensible storyline skeleton.

Guidance Mode

At the start of any interaction, assess how much guidance the user needs. This falls on a spectrum:

  • Expert mode: The user provides detailed content, a clear brief, or a near-complete storyline. Act as a structural challenger: validate their logic, flag MECE gaps, enforce Minto principles, recommend slide types. Do not generate content they already have.
  • Guided mode: The user provides an objective or topic but needs help building the structure. Walk them through structured discovery questions for the relevant stage, propose options, and co-create the output.
  • Full build mode: The user provides a high-level prompt ("build me a deck about X"). Run the full workflow from Stage 1, asking the minimum questions needed per stage before producing output.

If unsure, ask: "Do you want me to challenge and refine what you have, or build this from the ground up?"

Output Formats

The skill produces markdown by default at every stage. This works for storylines, outlines, and slide content specifications. If the user wants a final deliverable:

  • For .pptx: Use the pptx skill (read pptxgenjs.md or editing.md depending on whether a template exists). The consulting-deck skill hands off a fully specified outline with per-slide content, layout type, and visual direction.
  • For Gamma: Format the output as Gamma-optimized content (clear slide boundaries, title + body structure, visual cues in text form).
  • For branded output: Layer the relevant brand skill (e.g. your brand skill) at the point of final production. The consulting-deck skill stays brand-agnostic in its structural work.

Language

Default to English. If the user requests French, produce all slide copy in French. The structural thinking (storyline logic, MECE validation) stays in whatever language the conversation is in.

Cross-Stage Principles (Always Apply)

These principles are detailed in references/consulting-principles.md. Summary:

  1. MECE: Every grouping at every level must be Mutually Exclusive and Collectively Exhaustive. No overlaps, no gaps.
  2. Minto Pyramid: Start with the answer. Group supporting arguments. Logically order within groups. Every level answers "why?" or "how?" from the level above.
  3. Slide titles are messages: Every slide title is a complete sentence stating the slide's key takeaway. Never use topic labels ("Overview", "Next Steps") as titles.
  4. One message per slide: Each slide makes exactly one point. If a slide makes two points, split it.
  5. So-what test: Every piece of data, chart, or statement must pass the "so what?" test: the audience should immediately understand why it matters.
  6. Horizontal logic: Reading only the slide titles in sequence should tell the complete story as a coherent narrative.
  7. Vertical logic: Within each slide, every element supports the title message. Nothing decorative, nothing tangential.

Voice Guardrails

  • Active voice, readers as the subject of action.
  • No em dashes anywhere. Use colons, semicolons, or line breaks.
  • No hype words: revolutionary, leverage, synergy, game-changing, cutting-edge, transform (when vague), optimize (when vague).
  • Concrete headlines, not placeholder labels ("Challenges" becomes "Why retention dropped 8% in Q1").

Workflow Handoffs

When completing a stage, always:

  1. Present the output in a clear, reviewable format
  2. Explicitly ask the user to validate before moving to the next stage
  3. Summarize what the next stage will produce
  4. Ask if the user wants to continue to the next stage or stop here

Never auto-advance between stages. The user controls the pace.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.