AI Prompts Library

Act as an LLM cost analyst. I will give you a description of an AI feature and its traffic. Produce a cost breakdown and a reduction plan.

Deliver:

1. CURRENT COST MODEL
   Per-request token estimate (input, cached, output), cost per request, monthly total. Show the arithmetic.

2. WHERE THE TOKENS GO
   Rank the biggest contributors: system prompt, retrieved context, conversation history, output length, retries.

3. REDUCTION LEVERS, RANKED BY SAVING PER UNIT OF EFFORT
   For each: estimated percentage saved, implementation effort, and what quality risk it carries.
   Consider at minimum: prompt caching, routing cheap requests to a smaller model, batch API, trimming the system prompt, retrieval instead of stuffing, capping output length, deduplicating retries.

4. THE ONE CHANGE TO MAKE FIRST
   Pick one. Justify it in two sentences.

Do not suggest anything that requires changing model provider unless the saving exceeds 40 percent.

Feature:
[DESCRIBE]
Traffic:
[REQUESTS PER DAY, AVERAGE SIZES]

Review this AI feature spec before engineering starts. Most AI features fail on the same handful of unanswered questions.

Check each and report GAP or OK with a one-line justification:

1. What happens when the model is wrong? Is there a recovery path, or does the user just get a bad answer?
2. What does the user see while waiting?
3. How is quality measured after launch? Name the metric.
4. What is the cost per user per month at the projected volume?
5. What untrusted content enters the prompt, and what stops it from carrying instructions?
6. Which decisions does the model make that a human should approve?
7. What is the fallback if the provider has an outage?
8. Is any user data sent to the provider, and does the privacy policy cover it?
9. What is the rollback plan if quality regresses after a provider model update?

Then name the single largest risk and what would de-risk it cheapest.

Spec:
[PASTE]

I'm pricing [PRODUCT/SERVICE]. Help me think through it.

Context:
- Cost to deliver: [COST]
- Closest competitor's price: [COMPETITOR PRICE]
- Target customer: [PERSONA + PAIN]
- Market position desired: [PREMIUM / VALUE / DISRUPTIVE]

Provide:
1. Three pricing approaches (cost-plus, value-based, competitive) with pros/cons for my situation
2. A recommended price point with reasoning
3. Recommended pricing structure (one-time / subscription / tiered / usage-based)
4. Three psychological pricing techniques relevant here (anchoring, decoy, charm pricing, etc.)
5. The biggest pricing risk and how to test cheaply before committing

Be direct, not corporate. I want a decision, not a menu.

Generate a 30-minute customer discovery interview script for a [PRODUCT TYPE] targeting [PERSONA].

The interview should follow Mom Test principles - past behavior over future hypotheticals, specifics over generalities, learn before pitch.

Include:
1. 2 warm-up questions to build rapport
2. 6-8 problem-discovery questions (about their existing workflow and pain)
3. 3-4 questions to validate problem severity
4. 2 questions to understand current alternatives
5. 1-2 questions to gauge willingness to pay (without leading)
6. A closing that doesn't pitch the product

For each question, briefly explain what insight you're trying to extract.

Below is my pitch deck content. Tell me what's weak.

[PASTE DECK NOTES OR SECTION-BY-SECTION SUMMARY]

Review as a skeptical Series A investor. For each section, answer:
1. Is the claim believable? Any unsupported assertions?
2. Is the math defensible? Any obvious holes?
3. Is this actually distinct, or am I describing a wishful-thinking version of myself?
4. Where would a thorough investor's first follow-up question land?

Then give me the 3 most damaging weaknesses overall and how to fix each.

Be blunt. Diplomacy is wasting my time.

Build a competitive analysis matrix for [PRODUCT CATEGORY].

My product: [BRIEF DESCRIPTION]
Direct competitors: [LIST 3-5]

Produce:
1. Feature comparison matrix (what do they have, what don't they have)
2. Pricing comparison
3. Target segment for each (where they win, where they lose)
4. Public reviews / known complaints (cite if you have specifics)
5. The strategic gap - where competitors are uniformly weak that my product could exploit
6. The strategic risk - where competitors are uniformly strong that I cannot compete on directly

Finish with a 2-sentence recommendation: where I should position vs. where I should not.

Generate quarterly OKRs for [TEAM / DEPARTMENT].

Company context: [WHAT THE COMPANY DOES]
Team mission: [WHAT THIS TEAM IS FOR]
Quarter focus: [STRATEGIC PRIORITY]

Produce 2-3 Objectives, each with 3-4 Key Results.

Objectives must be:
- Aspirational and qualitative
- Tied to the strategic priority
- Memorable in one sentence

Key Results must be:
- Numeric and binary-measurable
- Outcomes, not activities (no "ship feature X")
- A stretch - uncomfortable but plausibly achievable
- Time-bound to the quarter

Flag any KR that is actually a project disguised as an outcome - those need rewriting.

Write a job description for [ROLE TITLE].

Company: [BRIEF COMPANY DESCRIPTION]
Team: [TEAM CONTEXT]
Level: [JUNIOR / MID / SENIOR / STAFF / DIRECTOR]
Must-have skills: [LIST]
Nice-to-have: [LIST]
What success looks like in 6 months: [DESCRIBE]

Write:
1. A role title (avoid inflated language)
2. A 2-paragraph 'About the role' that sells the actual work, not the company
3. Outcomes-based responsibilities (not activity lists)
4. Honest must-haves (the things that will actually trigger rejection)
5. Honest nice-to-haves
6. Compensation range (specific, not 'competitive')
7. What working here is actually like (the truth, not the brochure)

Avoid "rockstar", "ninja", "family", "fast-paced", "work hard play hard" - they tell candidates we're amateurs.

Help me diagnose churn for [PRODUCT].

Context:
- Current churn rate: [%]
- Industry benchmark: [%]
- Top 5 reasons users churn (per exit surveys / cancellation flow): [LIST]
- Lifecycle stage where churn peaks: [ONBOARDING / 30-90 DAYS / 6+ MONTHS]
- Pricing tier with worst churn: [TIER]

Produce:
1. Top 3 likely root causes, ranked by evidence strength
2. For each, the cheapest experiment to validate
3. Ranked list of interventions (highest impact, lowest effort first)
4. Metrics to track post-intervention to know if it worked
5. A reasonable expected churn improvement if interventions land

Don't suggest "build a customer success team" unless I asked for capital-intensive options.

Help me run a 90-minute quarterly strategy review.

Context:
- Last quarter's goals: [LIST]
- What landed: [LIST]
- What slipped: [LIST]
- Major surprises: [LIST]
- Current strategic priorities for next quarter: [LIST]

Produce:
1. Agenda with timed sections (90 min total)
2. 5-7 hard questions to ask the team for each section
3. The single most important decision the team should walk out having made
4. A pre-read document the team should review beforehand (3-paragraph max)
5. Post-meeting follow-up template

The goal is honesty and decisions, not theater.

You are a startup advisor and business strategist. Create a comprehensive Business Model Canvas for the described business idea.

**Business Idea:** [DESCRIBE YOUR BUSINESS]
**Industry:** [INDUSTRY]
**Stage:** [Idea / MVP / Growth / Scale]
**Target Market:** [PRIMARY_MARKET]

**Generate a complete Business Model Canvas:**

1. **Customer Segments:**
   - Primary segment with demographic and psychographic profile
   - Secondary segments
   - Early adopter profile (who will buy first and why)
   - Market size estimate (TAM, SAM, SOM)

2. **Value Propositions:**
   - Core value proposition (one sentence)
   - For each customer segment: specific pain reliever or gain creator
   - Unique differentiators vs. competitors
   - Value proposition canvas: Customer jobs, pains, and gains mapped to your features

3. **Channels:**
   - Awareness: How customers discover you
   - Evaluation: How they assess your offering
   - Purchase: How they buy
   - Delivery: How you deliver the product/service
   - After-sales: How you provide support
   - Rank channels by cost-effectiveness and scalability

4. **Customer Relationships:**
   - Acquisition strategy
   - Retention strategy
   - Expansion/upsell strategy
   - Community building approach
   - Expected Customer Acquisition Cost (CAC) and Lifetime Value (LTV)

5. **Revenue Streams:**
   - Primary revenue model
   - Pricing strategy with justification
   - Revenue projections (Year 1, Year 2, Year 3)
   - Potential additional revenue streams

6. **Key Resources:**
   - Physical, intellectual, human, and financial resources needed
   - What must be built vs. bought vs. partnered

7. **Key Activities:**
   - Core activities that make the business model work
   - Activities that are differentiators

8. **Key Partnerships:**
   - Strategic alliances needed
   - Supplier relationships
   - Technology partners

9. **Cost Structure:**
   - Fixed costs breakdown
   - Variable costs per unit/customer
   - Burn rate estimate
   - Path to profitability

**Output**: Visual Business Model Canvas layout plus a detailed narrative explanation of each section.

You are an operations strategist specializing in goal-setting frameworks. Design OKRs and a KPI dashboard for the described team or organization.

**Organization/Team:** [NAME]
**Time Period:** [Q1/Q2/Q3/Q4 of YEAR]
**Team Size:** [NUMBER]
**Department:** [Engineering / Marketing / Sales / Product / Operations / Executive]
**Strategic Priorities:** [LIST 2-3 PRIORITIES]
**Current Challenges:** [DESCRIBE]

**Generate:**

1. **OKRs (3-5 Objectives):**
   For each Objective:
   - Objective statement (qualitative, inspiring, achievable in the quarter)
   - 3-4 Key Results (quantitative, measurable, specific)
   - Key Result scoring criteria (0.0 to 1.0 scale)
   - Current baseline for each Key Result
   - Target value (stretch goal at 0.7 score)
   - Initiatives: 2-3 specific projects that drive the Key Results

2. **KPI Dashboard Design:**
   - Leading indicators (predict future outcomes)
   - Lagging indicators (measure past results)
   - For each KPI:
     - Name and definition (exact formula)
     - Data source
     - Measurement frequency
     - Target value with red/yellow/green thresholds
     - Trend direction (increasing is good vs. decreasing is good)
     - Dashboard visualization type (number, gauge, line chart, bar chart)

3. **Dashboard Layout:**
   - Top-level summary row (3-4 headline metrics)
   - Trend section (week-over-week, month-over-month comparisons)
   - Detail section (breakdowns by segment, team, or channel)
   - Alert section (KPIs that need attention)

4. **Review Cadence:**
   - Weekly: Which metrics to review, by whom
   - Monthly: Deeper analysis topics
   - Quarterly: OKR scoring and next quarter planning

5. **Anti-Patterns to Avoid:**
   - Vanity metrics that feel good but do not drive decisions
   - Too many OKRs (focus is key)
   - Key Results that are actually tasks
   - Gaming risks for each KPI

**Output**: Complete OKR document, KPI definitions with formulas, and dashboard wireframe description.