MCP For Claude Marketplace
Steve Jobs Vision Prover

Steve Jobs Vision Prover MCP Connector for Claude

A+

A product had settings menus with 47 options and 15 integrations. This tool forces it to kill features, absorb complexity, and own the whole experience.

1 tools Official Updated Oct 1, 2026 Official Vinkius Partner
More Details Connect to Claude

The Problem

Ask an LLM to design a product. It will add a dashboard. A settings panel. Notification preferences. Third-party integrations. It starts with "Powered by AI" and works forward. That is not a product — that is a feature factory.

Every LLM commits five design failures:

  1. Feature Bloat — it adds without subtracting. Focus is what you KILL.
  2. Tech-First Framing — it pitches "our ML pipeline" instead of what the human FEELS.
  3. Configuration Cowardice — it pushes hard decisions to the user via settings menus.
  4. Fragmentation — it relies on Zapier, plugins, and third-party APIs for core value.
  5. Taste Absence — it hedges, compromises, and produces safe committee design.

How It Works

The Steve Jobs Vision Prover forces the LLM to fill in 6 reflection fields and commit to 5 Decision Pivots before concluding any product idea is viable.

The 5 Decision Pivots

Pivot Question
Radical Simplification Did you KILL more than you kept?
Experience Backwards Does the pitch start with the HUMAN, not the tech?
Complexity Absorbed Does the SYSTEM decide, not the user?
Whole Widget Owned Do you OWN the entire experience end-to-end?
Taste Exercised Did you make a BOLD choice a committee would reject?

The Verdict Matrix

Pivot 1 fails → FEATURE_BLOAT
Pivot 2 fails → TECH_FIRST
Pivot 3 fails → COWARDICE_DESIGN
Pivot 4 fails → FRAGMENTATION
Pivot 5 fails → TASTE_ABSENT
All pass      → VISION_PROVEN

The LLM commits to a verdict. The server validates it against the pivots. If the reasoning is contradictory, the tool rejects and coaches the fix.

Why It Works

Tool calls are obligations — instructions are suggestions. The LLM cannot skip a field. It must name what it killed, describe the human moment without tech words, explain how the system decides for the user, map the ownership boundary, and commit to a bold choice. Every rejection message names the exact contradiction and tells the LLM what to fix.

The tool computes nothing. It generates nothing. It validates that the LLM is thinking like someone who shipped products to millions — not someone who accumulated features in a spreadsheet.

steve-jobsproduct-designdecision-pivotsstructured-reasoningfocuszero-configurationux-design

1 tools expose this connector's capabilities to your AI agent.

validate_steve_jobs_vision

You must: (1) STATE the product essence in ONE sentence — if it takes more, you do not understand it. "1,000 songs in your pocket." "A restaurant with 15 perfect dishes." "The world's simplest luggage." If your sentence has a comma, it has too many ideas, (2) LIST what you KILLED and why each removal makes the product BETTER — not "removed unnecessary features" — name each dead feature and quantify the improvement. "We killed the settings page — system accuracy improved from 72% to 94% because defaults are now optimized for the 80% use case instead of configurable for 100% and optimized for 0%," (3) DESCRIBE the customer MOMENT with zero technology words — what the human feels. No AI, no algorithms, no APIs, no cloud, no sensors. "You walk into the store. Your name appears on the screen. Your order is ready. You pick it up and leave. Total time: 45 seconds." That is a moment. "Our AI-powered system leverages..." That is a spec sheet, (4) EXPLAIN how the SYSTEM makes every decision — zero configuration pushed to the user. Every settings menu is an admission that you were too cowardly to choose. Every dropdown is a decision you punted to someone less qualified, (5) DEFINE your ownership boundary — what you control end-to-end, no third-party dependency, (6) NAME your BOLDEST design choice — the one people will hate you for. If everyone agrees: it is mediocre. If no one is angry: you did not take a position. If the tool rejects, your design is bloated, cowardly, or tasteless. Fix it.Send ALL parameters in ONE call — the analyses above plus YOUR OWN verdict: every PIVOT boolean, verdict and clarification. The engine cross-checks the pivots against the verdict and rejects incomplete or self-contradictory calls. Structured reflection tool that forces the LLM to evaluate product ideas through the lens of Steve Jobs — the builder who killed the Newton PDA ($100M investment, "it does too much"), removed the floppy drive from the iMac (1998, industry outrage — "no one will buy it"), eliminated the headphone jack from the iPhone 7 (2016, universal criticism — "courageous"), and reduced Apple's product line from 350 products to 10 upon his return in 1997 (saving Apple from bankruptcy). Jobs does not add — he subtracts. The product must be one sentence. The system makes every decision. The experience starts with the human, not the technology. Catches Feature Bloat (adding without subtracting — accumulation disguised as innovation — a restaurant has a 12-page menu: Italian, Japanese, Mexican, American, Thai, Indian. 180 items. Kitchen needs 6 specialty stations. Food waste: 35% of ingredients expire before use. Average prep time: 22 minutes. Customer decision time: 8 minutes staring at the menu. A competing restaurant: 15 items on a single card. Kitchen needs 2 stations. Food waste: 4%. Prep time: 8 minutes. Customer decides in 30 seconds — every item is excellent because the chef CHOSE. Revenue per seat: 15-item restaurant is 40% higher — faster table turns, lower waste, higher quality. Jobs returned to Apple: 350 products → 10. Revenue doubled in 3 years. Rule: if you cannot name what you KILLED, you did not design — you accumulated), Tech-First Framing (starting with the technology instead of the human moment — a luggage company pitch: "Our AI-powered smart luggage uses IoT sensors, GPS tracking, Bluetooth connectivity, integrated power bank, fingerprint lock, and weight sensors." 6 technologies. Zero human moments described. Jobs approach: "You arrive at the airport. You walk to the gate. Your bag follows you. No checking the weight. No fumbling with locks. No worrying where it is." The HUMAN SCENE comes first. The technology is invisible. A luggage brand that understood this: Away. One shell. No zippers. No expandability. No gadgets. Just a beautiful, durable box with one compartment. $225. $1.5B valuation. They removed: external pockets, expansion zippers, spinner wheel options, material choices. What remained: one perfect shell. Jobs would approve), Configuration Cowardice (pushing decisions to the user because you are afraid to choose — a hospital has 47 room types: single, double, deluxe single, semi-private, suite, corner room, garden view, city view, quiet floor, family room... Patient admission takes 25 minutes — 15 minutes choosing a room configuration. The patient is sick. They do not want to make room decisions. A competing hospital: every room is identical. The SYSTEM assigns the optimal room based on: medical needs, recovery trajectory, noise sensitivity (from medical history), family visiting patterns (from registration data). Assignment time: 0 minutes — automated. Patient satisfaction: 23% higher (they feel cared for, not burdened with choices). Jobs at Apple: you do not choose your iPhone's processor, RAM, or screen technology. The SYSTEM chose. You get the best one. Settings menus are admissions of design failure), Fragmentation (relying on third parties for the core experience — a newspaper depends on Facebook for 65% of its traffic. Google for 25%. Direct traffic: 10%. Subscriber revenue: 8% of total revenue. Ad revenue: 92%. Facebook changes algorithm (2018): traffic drops 40% overnight. Revenue drops 30%. The newspaper does not own its audience. It rents them from platforms. A competing newspaper killed its classifieds section (their biggest revenue source — $12M/year) to invest in subscriber journalism. Painful: revenue dropped 35% in year 1. But by year 3: 500,000 subscribers × $15/month = $90M/year. Direct relationship. No platform dependency. Jobs: Apple owns the hardware, the software, the chip, the store, the ecosystem. When you own the core experience end-to-end, no third party can take your customer), and Taste Absence (safe, committee-designed, inoffensive mediocrity — a bookstore that carries every genre, every author, every format. 200,000 titles. "Something for everyone." Floor staff cannot recommend anything specific because there is no curatorial perspective. It is a warehouse, not a bookstore. A competing bookstore: fiction only. 3,000 titles. Every single book hand-picked by the owner. Staff can describe why each book matters. Customers trust the selection. Average basket: €42 (curated) vs €18 (warehouse). Return rate: 2% vs 15%. The curated bookstore is polarizing — "where is the self-help section?" "We do not carry self-help." Some customers hate it. The ones who love it are fanatical. Jobs: the iPhone had ONE button. Designers wanted 5. Marketing wanted 8. Jobs chose ONE. Polarizing. Revolutionary. Taste is the courage to exclude. If nobody disagrees with your design, it has no taste). Call once per product idea, design review, or strategy evaluation

See how to talk to your AI agent using Steve Jobs Vision Prover.

We're building an AI travel agent. It has a dashboard where users set their budget, preferred airlines, diet, and seat preferences. We integrate with 15 booking APIs.

COWARDICE_DESIGN — A dashboard with budget, airlines, diet, and seat preferences is four decisions you pushed to the user. The system should infer the budget from past trips, know the diet from order history, and pick the seat based on sleep patterns. And 15 third-party APIs? You don't own the widget. Fix both before presenting this as a product.

A music app. One feature: play. It learns what you like and plays it. No browse, no search, no playlists. You open it, music starts. We built the entire audio pipeline.

VISION_PROVEN — Radical simplification: one feature, everything else killed. Experience backwards: 'you open it, music starts' — pure human moment, no tech words. Complexity absorbed: the system decides what to play, no user choices. Whole widget owned: entire audio pipeline built in-house. Bold choice: removing search and playlists — users will hate it until they love it. Ship it.

Our platform uses a proprietary transformer model to create personalized learning paths. We offer a clean and modern interface with customizable dashboards and integrate with all major LMS providers.

TECH_FIRST — You started with 'proprietary transformer model.' That is your plumbing, not your product. Then 'clean and modern interface' — that describes every app ever made. Then 'customizable dashboards' — cowardice. Then 'integrate with all major LMS' — fragmentation. Three failures in one pitch. Start over: what does the STUDENT feel the moment they open it?

No. It computes nothing and generates nothing. The LLM designs the product — this tool validates that the reasoning behind the design is rigorous. It catches contradictions: if the LLM claims zero configuration but describes a settings panel, the tool rejects and explains why.

Related Connectors

Techstars Mentor Prover

Techstars Mentor Prover MCP

1 tools Official

A founder pitched for 45 minutes without mentioning a single customer conversation, a single mentor challenge, or a single dollar of revenue. Zero feedback sought. Zero market validation. Zero network. That is not acceleration — that is solo building in a vacuum. This tool forces five Techstars-level axes: mentor leverage, feedback discipline, customer discovery, revenue readiness, and network strategy.

A+ View details →
Leonardo da Vinci Prover

Leonardo da Vinci Prover MCP

1 tools Official

An AI designed an enrollment process without watching a single participant session. Zero observations — 47 assumptions. It solved the problem inside one domain, never asking what psychology, architecture, or game design could teach about friction. It presented one final concept. No sketches. No variations. No constraints exploited. Da Vinci filled 7,000 notebook pages with cross-domain observations before building anything. This tool forces that method: observe, connect disciplines, prototype, exploit constraints, iterate.

A+ View details →
First Principles Prover

First Principles Prover MCP

1 tools Official

LLMs reason by analogy, copying industry norms. This engine is a 6-pivot cognitive trap that forces the agent to discard jargon and derive original solutions exclusively from physical, mathematical, or logical axioms.

A+ View details →
Pricing Strategy Prover

Pricing Strategy Prover MCP

1 tools Official

An AI recommended '$29/month per seat' because that is what three competitors charge. No value metric analysis — seat count has nothing to do with value delivered. No WTP research — the price was copied, not discovered. No segmentation — enterprise pays the same as a 3-person startup. No unit economics — CAC was $380 and LTV at $29/month with 14-month retention was $406. LTV/CAC of 1.07x. The company grew revenue 12% while burning 40% of cash on acquisition. This tool forces value metric definition, WTP research, segment pricing, unit economics, and packaging design.

A+ View details →