sendaifun/solana-new/create-pitch-deck
Claude Code and agent skill by sendaifun/solana-new
- community
Checking that the SKILL.md file is still in the repository.
Create a structured pitch deck for a crypto project. Use when a user says "create a pitch deck", "help me pitch", "I need slides", "prepare for demo day", "investor presentation", or "grant application". Reads idea-context.md and build-context.md from prior phases if available.
Install
npx vskill@latest install sendaifun/solana-new/create-pitch-deckSKILL.md
Preamble (run first)
If TEL_PROMPTED is no: Before starting the skill workflow, ask the user about telemetry. Use AskUserQuestion:
Help superstack get better! We track which skills get used and how long they take —no code, no file paths, no PII. Change anytime in~/.superstack/config.json.
Options:
- A) Sure, help superstack improve (anonymous)
- B) No thanks
If A: run this bash:
If B: run this bash:
This only happens once. If TEL_PROMPTED is yes, skip this entirely and proceed to the skill workflow.
Wrong skill? See SKILL_ROUTER.md for all available skills.
Create Pitch Deck
Overview
Generate a compelling, investor-ready pitch deck using proven frameworks from Sequoia, YC, a16z, and Guy Kawasaki — adapted for crypto/Solana projects. Produces a polished HTML artifact with slide content, speaking notes, narrative arc, and audience-specific tailoring.
This skill doesn't just generate slides — it coaches you through building a narrative that lands.
Core Principles
Hackathon judges review hundreds of submissions quickly and often skim for the essentials. Capture attention in the first 30 seconds with a punchy one-liner using a sharp, relatable analogy that instantly clarifies the idea — "HIP-3 for Solana", "Jupiter for lending", "Uber for compute". Follow immediately with a crisp, high-quality demo that shows the working product in action. Structure the deck as a simple narrative: clearly explain the problem, the innovative solution, and the business model or path to users. Skip stiff slide templates — focus on storytelling. Keep everything short, visually polished, aesthetic, and easy to scan. Draw inspiration from top competitors on design and flow. Clarity and impact win every time.
Non-Negotiables (Read First)
- Never skip the interview. Don't auto-generate from context files alone. Every pitch needs a human conversation to find the real story.
- Stay blunt. If their "problem" is vague, push back. If their "why now" is weak, say so. If they have no traction, don't hide it — help them frame what they DO have.
- Problem first, always. No one cares about Solana until they care about the problem. (Sequoia rule #1)
- One idea per slide. If you need two slides, use two slides. (YC rule)
- Real numbers only. "Growing fast" is not a metric. Mark unknowns as "TBD".
- Crypto necessity must be proven. "Why blockchain?" must have a concrete answer. If removing the blockchain doesn't break the product, the pitch has a fatal flaw — say so.
- The ask must be specific. Amount + instrument + use + timeline + milestone. "We're raising" is not an ask.
- Always generate the HTML artifact. Using the design system from references/deck-design-system.md and slide templates from references/slide-templates.md.
- 3 questions max per message. Conversational rounds, not a questionnaire dump.
Workflow
Phase 1: Context Gathering (Silent)
Before asking anything, read what exists:
.superstack/idea-context.md— Product concept, audience, value prop.superstack/build-context.md— Tech stack, features, deployment status- Output from
competitive-landscapeskill,defillama-researchskill if available
Also auto-detect project assets:
If you find useful context, mention it during the interview: "I see from your idea-context that you're building X for Y — let's start from there."
Phase 1.5: Live Research (mandatory before deck generation)
Every deck needs real, current numbers. Never use stale data from context files or training data alone. Before generating any slides, run a research pass for the specific idea:
Market data:
- Use DefiLlama MCP or web search to pull current TVL, volume, and protocol stats for the relevant category
- Get competitor TVL/volume/user numbers as of today, not from memory
- Pull total market size data (perps volume, lending TVL, stablecoin market cap, etc.)
Competitor landscape:
- Identify the 3-5 closest competitors and get their current metrics
- Check what's launched in the last 3 months in this space
- Note any recent funding rounds, partnerships, or shutdowns
Ecosystem data:
- Solana-specific stats: total TVL, DEX volume, number of active programs, relevant Pyth feeds
- Any recent ecosystem announcements or regulatory changes that affect the idea
How to research:
- Use web search, DefiLlama MCP, CoinGecko MCP, or any available data tools
- If MCP tools are unavailable, use web search as fallback
- Cross-reference at least 2 sources for key numbers
- Mark any numbers you couldn't verify as "~estimate" in the deck
Every number in the deck must come from this research pass, not from the interview or context files alone. If a stat is outdated or unverifiable, either update it or mark it clearly.
Check for recent hacks/exploits. Before citing any protocol as a competitor, integration, or reference point, verify it hasn't been recently compromised. Don't reference hacked protocols positively in the deck.
Phase 2: Deep Interview
This is a conversation, not a form. Start with anchor questions, pull deeper based on answers. Do NOT move to deck generation until you can clearly articulate: the problem, the solution, the proof, the audience, and the ask.
Round 1 — The Story (ask all three together):
- "Explain what you do to me like I'm someone you met at a bar — no jargon."
Why: Forces plain language. If they can't say it simply, they don't understand it yet (YC principle). The bar test is the first line of the title slide.
- "What personal experience made you angry enough to build this?"
Why: Founder-problem fit is the #1 signal VCs look for at pre-seed. (a16z: "We back founders who have lived the problem.") If they didn't live it, ask "then who told you about this problem, and why did you listen?"
- "What's the most surprising thing a user has told you?"
Why: Reveals customer discovery depth. If they haven't talked to users, that's a red flag — but a fixable one. Push: "Have you talked to potential users? What did they say?"
Round 2 — The Evidence (adapt based on Round 1):
- "What's your single strongest proof that this works?"
Why: Forces prioritization. Could be: users (DAU), money (revenue/TVL), speed (testnet metrics), demand (waitlist), or conviction (LOIs/pilots). Follow-up if weak: "If you have nothing yet, what's the cheapest experiment that would prove demand in 2 weeks?"
- "What changed in the last 12 months that makes this possible NOW?"
Why: Sequoia's most overlooked slide. "Why Now" separates ideas from opportunities. Common catalysts: new regulation (MiCA), tech breakthrough (state compression), market shift (institutional on-ramps), behavior change (mobile-first wallets). Push back if generic: "'Crypto is growing' is not a why-now. What specific event or change created this window?"
- "If I ripped out the blockchain from your product, what breaks?"
Why: The blockchain necessity test. If the answer is "nothing" — the pitch has a fatal flaw. Help them find the real crypto angle or be honest that it's weak. Good answers: "Composability breaks — other protocols can't build on us." / "Permissionless access breaks — no bank account needed." / "Trustless settlement breaks — agents can pay each other without intermediaries."
Round 3 — The Audience & Ask (ask based on Round 2):
- "Who exactly will you be presenting this to?"
Why: Everything changes based on audience. Push for specificity:
- Hackathon judges → Show working demo, innovation, completeness
- VCs → TAM/SAM, revenue model, team, defensibility
- Grant committees → Ecosystem impact, open-source, Solana-specific value
- Accelerators → Execution speed, coachability, early signals
- Strategic partners → Integration value, mutual benefit, user overlap
- "What's the specific outcome you want? Be precise."
Why: "Raise money" is not specific. Push for: "$500K SAFE at $5M cap to hire 2 engineers and reach 1K DAU in 6 months." OR: "Win the DeFi track at Colosseum Radar."
- "What's the question you're most afraid someone will ask?"
Why: This reveals the biggest weakness. If they say "I don't know" — probe: "Is it about traction? Team size? Why crypto? Competition?" The answer shapes the objection prep.
Round 4 — Taste & Positioning (ask if needed):
- "Do you have a preferred writing tone? Some folks want lowercase and casual, others want polished and corporate."
Why: The tone of slide text, speaking notes, and social copy should match the founder's voice — not sound AI-generated. Read ../tone-guide.md for defaults if they have no preference.
- "Have you seen a pitch or deck you loved? Any industry."
Why: Reveals their quality bar and communication style. If they name something, use it as a reference. If not, offer: "Clean and minimal like Airbnb's 10-slider? Data-heavy like Coinbase's seed deck? Story-driven like Buffer?"
- "Do you have brand colors? If not, what's the vibe — corporate and trustworthy, or bold and consumer?"
Why: NEVER default to purple or AI-ish gradients. The palette must come from the user's brand or be intentionally chosen. See references/deck-design-system.md for recommended palettes (Navy+White for DeFi, Teal+Ivory for consumer, Black+White for premium). Ask, don't guess.
- "Who are the 2-3 competitors someone will mention? What's your honest take on them?"
Why: Saying "no competition" kills credibility. Competitors include: direct rivals, indirect alternatives, and "doing nothing." (Peter Thiel: the biggest competitor is always the status quo.)
Interview Rules
This is the start of the file. Read the full SKILL.md on GitHub.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/launch/create-pitch-deck/SKILL.md
Popularity
- Repository stars
- 164
- Repository forks
- 46
- 7-day trend score
- 26
Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 8 Aug 2026, 2 months ago.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/sendaifun/solana-new/create-pitch-deck)