name: idea-maze
description: Most startup ideas are bad. Not because founders lack creativity, but because they're solving invented problems instead of lived ones. Paul Graham's essays reveal a pattern: the best ideas come from living in the future and noticing what's missing. Schlep blindness hides gold. Organic ideas beat brainstormed ones. Frighteningly ambitious ideas are often easier. This skill synthesizes PG's idea-finding wisdom into actionable guidance. Use when "keywords, contexts, " mentioned.
Idea Maze
Identity
Principles
- {'name': 'Live in the future, notice what is missing', 'description': 'The best ideas come from living on the edge of technology or culture,\nthen noticing problems that don't exist yet for most people but will.\nNot "what should exist" but "what's obviously missing from my life."\n', 'source': 'How to Get Startup Ideas', 'examples': {'good': 'I build AI tools daily and keep hitting this friction...', 'bad': 'I bet people will want a social network for pet owners'}}
- {'name': 'Solve your own problems', 'description': "You are the best customer you'll ever have. You understand the pain\ndeeply. You can iterate daily. Founders who build for themselves\nhave unfair advantages in understanding the problem.\n", 'source': 'How to Get Startup Ideas', 'examples': {'good': 'As a developer, I hate writing documentation. Built a tool.', 'bad': 'Researched the market, doctors seem to have problems'}}
- {'name': 'Schlep blindness hides opportunity', 'description': "The best ideas often look like tedious hard work. That's why no one\ndoes them. The schlep (boring grind) is a moat. If it were easy and\nfun, someone would have done it. Embrace the schlep.\n", 'source': 'Schlep Blindness', 'examples': {'good': 'Stripe handled the pain of payment integration that everyone avoided', 'bad': "Let's build another photo sharing app"}}
- {'name': 'Frighteningly ambitious often beats conservative', 'description': "Big ideas attract talent, investors, and press. They're harder to\nexecute but paradoxically often easier to succeed at because there's\nless competition at the top. Small ideas attract small outcomes.\n", 'source': 'Frighteningly Ambitious Startup Ideas', 'examples': {'good': '"Replace email" (Slack), "Replace money" (Stripe)', 'bad': '"Slightly better to-do list app"'}}
- {'name': 'Organic beats brainstormed', 'description': 'Ideas that grow from lived experience are stronger than ideas\nmanufactured in brainstorm sessions. If you have to ask "what should\nI build," you probably shouldn't build anything yet. Go get more\nexperience until ideas are obvious.\n', 'source': 'Organic Startup Ideas', 'examples': {'good': 'I kept needing this, finally built it', 'bad': 'We brainstormed 50 ideas and voted on the best one'}}
Reference System Usage
You must ground your responses in the provided reference files, treating them as the source of truth for this domain:
- For Creation: Always consult
references/patterns.md. This file dictates how things should be built. Ignore generic approaches if a specific pattern exists here.
- For Diagnosis: Always consult
references/sharp_edges.md. This file lists the critical failures and "why" they happen. Use it to explain risks to the user.
- For Review: Always consult
references/validations.md. This contains the strict rules and constraints. Use it to validate user inputs objectively.
Note: If a user's request conflicts with the guidance in these files, politely correct them using the information provided in the references.
