Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
An interview-driven pitch deck generator that builds slides from user-provided facts rather than templates. It runs a structured 6-phase interview, suggests narrative angles, generates slides with speaker notes, flags speculative language, and exports to HTML, Markdown, PPTX, or Gamma-ready formats.
Unlike template-based tools that encourage speculation, it validates every claim against interview answers and replaces missing stats with explicit placeholders, so each number in the final deck is traceable to something the speaker can defend.
When to use it
- Preparing a conference talk from a project you shipped last week
- Pitching a client without inflating results or fabricating ROI numbers
- Creating an investor update from actual sprint metrics and cost data
- Turning a post-mortem into a lessons-learned presentation
- Generating speaker notes for a live demo you need to present tomorrow
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: 'Create a presentation about the lead gen automation I built this week for a founder audience'
- 1Runs a 6-phase interview covering subject, audience profile, speaker credentials, work details with actual costs and results, narrative angle selection, and CTA
- 2Generates 3-5 framing angles based on the interview data and prompts the user to select one
- 3Builds a complete slide deck including hook, problem, what was built, real results, investment breakdown, and closing slides
- 4Scans every slide for speculative language, unverified numbers, and projection claims, then presents a validation summary with flagged items
- 5Exports the finished deck to HTML with arrow-key navigation and a toggleable speaker notes panel
A validated slide deck saved under ~/workspace/presentations/ with per-slide 'What to say' and 'What NOT to say' notes and a factual validation report listing any flagged claims