Engineered, not typed.
50 production-grade prompts for founders, CEOs, and business leaders. Every prompt follows Anthropic best-practice prompt engineering. Every output is structured, specific, and immediately usable.
Browse All Prompts · Decision Tree · How to Use · Download Excel · Contribute
Most prompt libraries are one-liners dressed up as resources. Type a sentence, get a sentence back. This is not that.
Every prompt in this collection is a complete system: explicit expert role assignment, XML tag structure, bracketed placeholder variables, exact output format specifications, numbered instruction steps, chain-of-thought reasoning for complex decisions, and a self-check evaluation block the model runs before delivering output. That is the full Anthropic prompt engineering stack, applied to every single prompt.
The result is not AI-flavoured output that needs two hours of cleanup. It is a structured, specific, immediately usable deliverable. The kind a Big 4 consultant would charge five figures to produce.
The difference is in the variables. Vague inputs produce vague outputs. Every prompt in this library tells you exactly what to enter, why it matters, and what a specific example looks like. Fill the brackets properly and the output is ready to use without editing.
The difference is in the constraints. Each prompt tells the model what it must never do: no invented statistics, no generic advice, no softened verdicts, no placeholder content. Constraints are what separate a prompt that produces something publishable from one that produces something plausible.
The difference is in the evaluation block. Before delivering output, the model checks its own work against five specific criteria. The prompt does not trust the model. It verifies.
This is not a list of ideas. It is a production-grade operating system for the highest-stakes decisions a founder makes.
| Role | How You Use This |
|---|---|
| Founders and CEOs | Validate ideas, build models, plan fundraises, design GTM, without hiring consultants for every decision |
| Startup operators | SOPs, hiring rubrics, vendor evaluations, weekly planning: the operational stack, automated |
| Consultants and advisors | Client-ready deliverables in minutes, not days |
| Students and early-career | Learn how senior practitioners think about every domain of business |
The prompts you need before you write a line of code or spend a dollar.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 01 | Startup Idea Validator | Essential | RAG verdict across problem, market, competition, founder-market fit + 3 validation experiments |
| 02 | Founder-Problem-Market Fit Analyser | Practitioner | Scores fit across 3 dimensions, names your single most dangerous knowledge gap |
| 03 | Business Model Canvas Builder | Essential | All 9 blocks with embedded assumptions and a weakest-block priority action |
| 04 | MVP Scope and Validation Designer | Practitioner | Assumption risk matrix + pre-build validation experiment design |
| 05 | Beachhead Market Selector | Advanced | Weighted scoring across 6 criteria with one unambiguous recommendation |
| 06 | One-Page Business Clarity Document | Essential | Six-section clarity document with honesty scores and improved alternatives |
Know your market better than anyone else in the room.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 07 | Market Sizing and Opportunity Analyser | Practitioner | TAM/SAM/SOM with documented assumptions + investor-ready narrative |
| 08 | Competitor Landscape Mapper | Practitioner | Direct, indirect, and status quo competitor profiles with positioning statement |
| 09 | Competitor Review Pain-Point Miner | Practitioner | Cross-competitor frustrations converted into 3 ready-to-use messaging angles |
| 10 | Market Gap and White Space Identifier | Advanced | Competitive clustering + pricing gaps mapped to a specific occupiable position |
Build something that lasts. Know why it will.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 11 | Business Model Canvas Builder | Essential | 9 blocks with assumptions + alternative model for better unit economics |
| 12 | Revenue Model Designer | Practitioner | 3 model types evaluated + full unit economics at recommended price point |
| 13 | Competitive Moat Analyser | Practitioner | 6 moat types scored, stress-tested against a named attacker |
| 14 | Business Model Devil's Advocate | Advanced | Hidden assumptions, structural failure points, VC bear case |
| 15 | Pivot vs Persevere Framework | Advanced | Signal/noise classification + direct Persevere, Pivot, or Stop recommendation |
| 16 | Unfair Advantage Finder | Essential | Advantages inventoried, scored, with exploitation strategies |
The numbers that matter. Calculated properly.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 17 | Startup Budget and Cost Builder | Essential | Month-by-month burn table + break-even revenue as a sales target |
| 18 | Unit Economics Analyser | Practitioner | Full CAC, LTV, payback with Series A benchmarking |
| 19 | Burn Rate and Runway Planner | Practitioner | 3 scenarios + trigger points for raise/cut/cash-flow-positive |
| 20 | Revenue Forecast Builder | Advanced | Bottom-up 12-month ARR build + 3-year projection with assumption register |
Raise on your terms. Walk in prepared.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 21 | Funding Route Evaluator | Essential | All viable routes scored on 4 dimensions + 3 pre-approach actions |
| 22 | Pitch Deck Narrative Architect | Practitioner | Slide-by-slide script: key message, visual direction, speaker notes |
| 23 | Investor Targeting Brief | Practitioner | 3-tier investor list + personalisation angles + cold email template |
| 24 | Investor Update Template | Practitioner | Structured update: status, metrics, challenges with responses, specific asks |
| 25 | Due Diligence Preparation Pack | Advanced | 6-category DD checklist with severity ratings + preparation timeline |
Get to market fast. Get it right first time.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 26 | Ideal Customer Profile Builder | Practitioner | Firmographic ICP + buyer personas + trigger events + disqualifiers |
| 27 | GTM Strategy and Channel-Market Fit | Advanced | 6 channels scored + primary motion designed + 30/60/90-day signals |
| 28 | Launch Playbook Builder | Practitioner | Pre-launch, launch day, post-launch checklists with owners and definitions of done |
Sound different. Be remembered.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 29 | Brand Voice and Messaging Guide | Practitioner | Voice principles + message hierarchy + 10 before/after copy examples |
| 30 | Content Strategy and Calendar | Practitioner | 3 differentiated pillars + 90-day calendar + adjustment triggers |
| 31 | Thought Leadership Architect | Practitioner | Publish-ready piece + 3 alternative hooks + quality scoring |
| 32 | Personal Brand Positioning | Essential | Brand position + LinkedIn About + cross-platform messaging + 12-week plan |
Close more. Discount less. Build a machine.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 33 | Sales Funnel Architect | Practitioner | Stages as buyer commitments + conversion metrics + impact-projected improvements |
| 34 | Cold Outreach Sequence Builder | Practitioner | Multi-touch sequence + trigger-event variants + follow-up decision tree |
| 35 | Objection Handler and Battle Card | Practitioner | Acknowledge-reframe-evidence responses + one-page mid-call battle card |
| 36 | Pricing and Negotiation Playbook | Advanced | Concession framework + walk-away points + 3 negotiation scripts as dialogue |
Build the right thing. Build it right.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 37 | Product Requirements Document Writer | Practitioner | Binary acceptance criteria + success metrics + explicit out-of-scope |
| 38 | Feature Prioritisation and Roadmap | Practitioner | Weighted scoring across 5 dimensions + phased roadmap by team capacity |
| 39 | Build vs Buy vs Partner Framework | Advanced | Scored evaluation + reversibility assessment + single clear recommendation |
Systems that scale without you.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 40 | SOP Architect | Essential | Single-action steps + binary definitions of done + failure recovery procedures |
| 41 | Workflow Optimisation and Bottleneck Identifier | Practitioner | Root cause classification + optimised workflow with projected cycle time |
| 42 | Vendor and Tool Stack Evaluator | Practitioner | Must-have filter + weighted scoring + recommendation + second-place scenario |
Hire the right people. Build the right culture.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 43 | Job Description and Role Architect | Essential | 30/90/180-day milestones + observable requirements + culture statements that filter |
| 44 | Interview Scorecard and Hiring Rubric | Practitioner | STAR question bank + observable scoring rubric + red flag indicators |
| 45 | Culture and Values Blueprint | Practitioner | Behaviours + anti-behaviours + decision principles + anti-patterns map |
Know what you need. Know what it costs when you skip it.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 46 | Startup Legal Checklist | Essential | 5-category checklist with urgency ratings + pre-fundraise readiness |
| 47 | Founder Agreement Clause Generator | Practitioner | Clause library + plain-English explanations + negotiation points |
Think clearly. Decide well. Build deliberately.
| # | Prompt | Level | What It Produces |
|---|---|---|---|
| 48 | High-Stakes Decision Maker | Practitioner | Criteria-weighted matrix + pre-mortem + decision record with review trigger |
| 49 | Delegation Matrix and Time Audit | Essential | 2x2 categorisation + delegation plan + under-investment analysis |
| 50 | Weekly Founder Reset and Priority Planner | Practitioner | Blocker type classification + 3 goal-anchored priorities + energy-managed plan |
Not sure which prompt fits your situation? The Decision Tree maps every founder situation to the right prompt in 30 seconds.
Every prompt in this library follows the same structure. Copy. Fill. Run.
1. Find your situation in the Decision Tree or browse the table above
2. Copy the full prompt text from the .md file
3. Replace all [BRACKETED_VARIABLES] with your specific context
4. Paste into Claude, GPT-4o, Gemini, or any capable LLM
5. Get a structured, specific, immediately usable output
Vague variables produce generic outputs. Specific variables produce outputs you can use immediately.
| Variable | Generic: produces generic output | Specific: produces usable output |
|---|---|---|
[TARGET_CUSTOMER] |
B2B SaaS companies | CFOs at UK-regulated financial services firms with 500-2000 employees |
[COMPETITOR] |
Salesforce | Salesforce Sales Cloud, mid-market tier, £75k ACV |
[DECISION_TO_MAKE] |
Whether to hire | Whether to hire a full-time Head of Sales at £95k or promote the current SDR and hire two junior reps |
[PRODUCT_OR_SERVICE] |
Compliance software | FCA-regulated compliance monitoring SaaS that cuts quarterly reporting from 3 days to 4 hours |
The difference between a prompt that produces something publishable and one that produces something generic is not the prompt. It is the variables you put into it.
Every prompt uses Anthropic's best-practice engineering framework:
<role>
Specific expert role with domain experience
</role>
<context>
[YOUR_VARIABLES] , fill these in before running
</context>
<instructions>
Step 1: ...
Step 2: ...
Step N: Compile into the output format below
</instructions>
<output_format>
Exact specification of tables, sections, columns, scoring scales
</output_format>
<constraints>
What to avoid, quality guardrails, specificity requirements
</constraints>
<evaluation_criteria>
Self-check the AI runs before delivering output
</evaluation_criteria>| Level | Who It's For |
|---|---|
| Essential | Any founder, any stage. No prior knowledge needed. Run it today. |
| Practitioner | You understand the domain and want structured depth. |
| Advanced | Complex situations. Multi-step analysis. Chain-of-thought reasoning. |
These prompts are tested and optimised for Claude (recommended) but work with any major LLM:
| Model | Performance | Notes |
|---|---|---|
| Claude 3.5+ / Claude 4 | ⭐⭐⭐⭐⭐ | Best results. Follows XML structure precisely. |
| GPT-4o | ⭐⭐⭐⭐⭐ | Excellent. Full XML support. |
| Gemini 1.5 Pro / 2.0 | ⭐⭐⭐⭐ | Very good. Minor formatting variations. |
| Llama 3.1 70B+ | ⭐⭐⭐⭐ | Good. May need slight output format nudging. |
| Mistral Large | ⭐⭐⭐ | Solid on Essential/Practitioner. Advanced prompts may need tweaking. |
The complete library is available as a premium Excel file with:
- Cover page and full Contents index
- Decision Tree: find the right prompt for your situation in 30 seconds
- Individual sheets for all 50 prompts with variables, prompt, why-it-works, and sample output
- Domain colour coding across all 13 domains
- Internal navigation: every prompt links back to Contents and Decision Tree
Download the Excel File · Free · All 50 prompts · One file · No sign-up required
The Excel file includes the full prompt library with a Cover page, Contents index, Decision Tree, and individual sheets for all 50 prompts , each with variables table, the prompt, why-it-works, and sample output. Navigate between prompts with one click.
ai-prompt-library/
│
├── README.md ← You are here
├── assets/
│ └── banner.svg ← Repo banner
│
└── prompts/
├── 01-business-model-foundations/
│ ├── 01-startup-idea-validator.md
│ ├── 02-founder-problem-market-fit.md
│ └── ...
├── 02-market-research/
├── 03-business-model-strategy/
├── 04-financial-planning/
├── 05-fundraising-pitch/
├── 06-go-to-market/
├── 07-marketing-brand/
├── 08-sales-revenue/
├── 09-product-engineering/
├── 10-operations/
├── 11-people-culture/
├── 12-legal-compliance/
└── 13-leadership-productivity/
Every prompt in this library meets these standards:
- Role specificity: Not "business consultant" but "SaaS pricing strategist with 15 years of B2B subscription experience"
- XML tag structure:
<role>,<context>,<instructions>,<output_format>,<constraints>,<evaluation_criteria> - Placeholder variables: Every user input is a named
[BRACKETED_VARIABLE]with a description and example - Exact output format: Tables, scoring scales, section headers, word counts , no ambiguity in what the output looks like
- Evaluation criteria: The AI self-checks its output before delivering
- Chain-of-thought: Advanced prompts require step-by-step reasoning before producing output
- British English: Throughout. Organise, not organize. Behaviour, not behavior.
- Zero filler phrases: No "it is worth noting", no "in today's rapidly evolving landscape", no em dashes
Found a prompt that could be better? Have a domain that's missing? Contributions welcome.
To contribute a new prompt:
- Fork the repository
- Create a new branch:
git checkout -b prompt/your-prompt-name - Copy the prompt template and build your prompt
- Submit a pull request with the prompt title, domain, level, and a sample output
Quality bar for contributions:
- Must use the full XML structure
- Must have at least 4 placeholder variables
- Must include a specific, non-generic sample output
- Must pass the evaluation criteria self-check
- Must follow British English
MIT Licence. Use freely, commercially or personally. Attribution appreciated but not required.
Built on Anthropic prompt engineering principles.
Built by Kunal RK
I build high-quality, practical tools for founders and professionals. This library is completely free. No paywall. No sign-up. Just use it.
If this library saves you time or helps you think more clearly about your business, a ⭐ goes a long way.
We Can Just Build Things.