talk-preparation
Prepare a talk from scratch: audience analysis, time boxing per section, outline structure, speaker notes, rehearsal strategy, Q&A preparation, and nervous system management. Use for conference talks, team presentations, demos, and webinars of any length.
Best use case
talk-preparation is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Prepare a talk from scratch: audience analysis, time boxing per section, outline structure, speaker notes, rehearsal strategy, Q&A preparation, and nervous system management. Use for conference talks, team presentations, demos, and webinars of any length.
Teams using talk-preparation should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/talk-preparation/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How talk-preparation Compares
| Feature / Agent | talk-preparation | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Prepare a talk from scratch: audience analysis, time boxing per section, outline structure, speaker notes, rehearsal strategy, Q&A preparation, and nervous system management. Use for conference talks, team presentations, demos, and webinars of any length.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Talk Preparation Skill ## When to Activate - Preparing a conference talk, webinar, meetup, or team presentation - Time-boxing a talk outline to fit a slot - Writing speaker notes that aid delivery without being read verbatim - Preparing for Q&A with difficult questions - Rehearsing effectively within available time - Adapting a talk for a new audience or conference format --- ## Audience Analysis Before writing a single slide, answer these four questions: 1. **Who is in the room?** (role, technical level, familiarity with the topic) 2. **What do they already know?** (baseline — don't explain what they know, don't assume what they don't) 3. **What problem do they have that my talk addresses?** (if no answer, reconsider the talk) 4. **What should they do or think differently after this talk?** (the one outcome) Write the answer to question 4 in one sentence — this becomes the talk's thesis. --- ## Time Boxing ### Rules of thumb | Talk type | Pace | Buffer | |-----------|------|--------| | Technical / code-heavy | 1 slide per 2 min | +10% | | Keynote / story-driven | 1 slide per 3 min | +10% | | Workshop / interactive | 1 slide per 5 min | +15% | **Live runs 15% slower than rehearsal.** Budget for it. ### Time allocation per section For a 30-minute talk: ``` Opening hook: 2 min (no agenda slide) Context / setup: 4 min Main point 1: 7 min Main point 2: 7 min Main point 3: 5 min Conclusion + CTA: 3 min Buffer: 2 min ───────────────────────────── Total: 30 min ``` Adjust ratio to content — demo-heavy talks may spend 60% on one section. ### Slide count calculator ``` Slide count ≈ Talk duration (min) / Pace (min per slide) Example: 30 min technical talk = 30 / 2 = 15 slides ``` --- ## Outline Structure A talk outline has five components: ``` Talk: [Title] Audience: [Role, level] Duration: [X] minutes Thesis: [One sentence — what the audience will think differently after] Opening (X min) Hook: [Specific question, stat, or story — write it out in full] Why this matters to them: [explicit audience relevance] Section 1: [Name] (X min) Key point: [One sentence] Supporting evidence: [Data, example, demo] Transition to next: [Exact bridging sentence] Section 2: [Name] (X min) [same structure] Section 3: [Name] (X min) [same structure] Closing (X min) Takeaway 1: [Most important insight] Takeaway 2: [Second insight] Takeaway 3: [Third insight] CTA: [One specific action] Q&A Prep [See below] ``` --- ## Speaker Notes Format Speaker notes exist to prevent going blank — not to be read aloud. ### Format rules ``` WRONG — prose that gets read verbatim: "In this section I will explain how the system works by walking you through the architecture diagram and then I will show you the key components including the API gateway and the message queue." CORRECT — keywords + timing markers: [2 min] → gateway → queue → show diagram → "this is the part that surprised us" [pause for effect] → demo: open terminal, run `make demo` → wait for output → transition: "so how does this affect you?" ``` ### Include in notes - **Timing markers**: `[1:30]` at points where you need to be at that time - **Demo steps**: exact commands to run, in order - **Transition sentences**: write the bridging sentence to the next slide verbatim — transitions are hardest to improvise - **Pause indicators**: `[pause]` or `[breath]` after key points - **Fallback**: if demo fails, which screenshot shows the same thing? --- ## Rehearsal Strategy Three rehearsal passes, in order: ### Pass 1 — Out loud, with timer - Speak every word out loud (do not "think through" silently) - Set a visible timer - Do not stop for mistakes — push through - Note sections that ran long or where you stumbled ### Pass 2 — Record video - Record yourself on your phone or laptop camera - Watch it back at 1.25x speed — you will spot filler words, pacing issues, and awkward transitions - Fix the top 3 issues, then do another out-loud run ### Pass 3 — Live audience - Present to at least one real person (colleague, friend) - Ask them to note: anything confusing, anything slow, anything they'd ask a question about - Their confusion points are your Q&A list ### Time check guideline After Pass 1: if you ran >10% over, cut content. If >20% under, add depth or slow down — rushing is a sign of nerves. --- ## Q&A Preparation Generate 10 questions before the talk — you will get most of them. ### Question categories to cover 1. **"Why not X?"** — Why did you choose this approach over the obvious alternative? 2. **"What about scale?"** — Does this work at 10x the load? 3. **"How long did it take?"** — Budget / effort reality check 4. **"What failed?"** — What did you try that didn't work? 5. **"Is this open source / available?"** — Availability and access 6. **"How do you handle [edge case]?"** — One specific edge case relevant to the domain 7. **"What would you do differently?"** — Honest retrospective 8. **"Who is using this in production?"** — Proof of real-world use 9. **[Domain-specific technical question]** — Deepest technical question possible 10. **[Hostile / skeptical question]** — "This is just [competitor] but worse" ### Handling hostile questions ``` Pattern: Acknowledge → Reframe → Redirect "That's a fair concern. [Acknowledge the validity of the pushback.] In our context, [explain your specific constraint or assumption]. We've found [your evidence]. Happy to discuss more after the talk." ``` ### "I don't know" It is better to say "I don't know — I'll find out and post on the conference Slack" than to guess and be wrong publicly. Prepare this exact phrase. --- ## Checklist - [ ] Audience analysis written: who, baseline, problem, one outcome - [ ] Thesis stated in one sentence - [ ] Time-boxed outline with per-section minutes - [ ] Slide count matches pace rule (not more than 1 slide/2min for technical) - [ ] Opening hook written in full (not "I'll figure out something catchy") - [ ] Transition sentences written for every section hand-off - [ ] Speaker notes are keywords + timing markers, not prose - [ ] Demo fallback screenshot ready - [ ] Pass 1 (out loud, timed) completed - [ ] Pass 2 (video recording) completed - [ ] Pass 3 (live audience) completed - [ ] 10 Q&A questions prepared with answer bullets - [ ] "I don't know" phrase rehearsed
Related Skills
zero-trust-patterns
Zero-Trust security patterns — mTLS between microservices (Istio/SPIFFE), SPIRE workload identity, OPA/Envoy authorization, NetworkPolicy default-deny-all, short-lived credentials, service mesh security, and Kubernetes RBAC hardening.
wireframing
Wireframing and prototyping workflow: fidelity levels (lo-fi sketch → mid-fi wireframe → hi-fi prototype), tool selection (Figma, Excalidraw, Balsamiq), user flow diagrams, wireframe annotation standards, information architecture (IA) mapping, and the handoff from wireframe to visual design. For developers who need to communicate UI structure before writing code.
webrtc-patterns
WebRTC patterns — peer connection setup, ICE/STUN/TURN configuration, signaling server design, SFU vs mesh topology, screen sharing, media track management, and reconnect/ICE restart handling.
webhook-patterns
Webhook patterns for receiving, verifying (HMAC), and idempotently processing third-party events. Covers Stripe, GitHub, and generic webhook patterns, delivery guarantees, retry handling, and testing.
web-performance
Web performance optimization: Core Web Vitals (LCP, CLS, INP), Lighthouse CI with budget configuration, bundle analysis (webpack-bundle-analyzer, vite-bundle-visualizer), hydration performance, network waterfall reading, image optimization (WebP/AVIF, srcset), and font performance.
wasm-performance
WebAssembly performance: wasm-opt binary optimization, size reduction (panic=abort, LTO, strip), profiling WASM in Chrome DevTools, memory management (linear memory, avoiding GC pressure), SIMD, and multi-threading with SharedArrayBuffer.
wasm-patterns
WebAssembly patterns: wasm-pack, wasm-bindgen (JS↔Wasm interop), WASI, Component Model, wasm-opt, Rust-to-WASM compilation, JS integration (web workers, streaming instantiation), and production deployment (CDN, Content-Type headers).
visual-testing
Visual Regression Testing: tool comparison (Chromatic/Percy/Playwright screenshots/BackstopJS), pixel-diff vs AI-based comparison, baseline management, flakiness strategies (masks, tolerances, waitForLoadState), CI integration with GitHub Actions, and Storybook integration.
visual-identity
Brand identity development: color palette construction (primary/secondary/semantic/neutral), logo concept brief writing, typeface pairings, brand voice definition, mood board direction, and Brand Guidelines document structure. Use when establishing or evolving a visual brand — not for implementing existing tokens.
ux-micro-patterns
UX micro-patterns for every product state: Empty States, Loading States (skeleton screens, spinners, optimistic UI), Error States, Success States, Confirmation Dialogs, Onboarding Flows, and Progressive Disclosure. These patterns apply to every feature — done wrong, they're the biggest source of user confusion.
typography-design
Typography as a creative discipline: typeface selection criteria, type pairing (serif + sans, display + body), modular scale systems, line-height and tracking ratios, hierarchy construction, and web/mobile rendering considerations. The decisions behind design tokens, not the tokens themselves.
typescript-testing
TypeScript testing patterns: Vitest for unit/integration, Playwright for E2E, MSW for API mocking, Testing Library for React components. Core TDD methodology for TypeScript/JavaScript projects.