| name | startup-explorer |
| description | Pre-validation startup idea explorer — bridges career expertise to a testable startup hypothesis. Triggers when a user wants to explore startup ideas, find a customer niche, discover what to build, identify product-market fit opportunities, or turn domain expertise into a business concept. Also triggers on: "startup idea", "what should I build", "find my niche", "explore business ideas", "side project to startup", "monetize my expertise", "customer discovery", "who would pay for this", "problem worth solving". Reads from career-advisor data. Feeds into idea-validation and business-model skills.
|
Startup Explorer
Go from "I'm good at X" to "here's a testable startup hypothesis with a specific customer" — before writing a line of code or a pitch deck.
Core Principle
"The best startup ideas come from expertise you already have, aimed at pain you've already witnessed."
Most first-time founders fail because they start with a solution ("I'll build an app for X") instead of a problem they deeply understand. This skill mines YOUR career for problems worth solving and customers you already know how to reach.
Prerequisites: Career Data
Read from the career-advisor folder. Detect career root:
Check in order:
1. my_knowledge/career/ (preferred)
2. knowledge/career/ (alternative)
If not found: "Run /career-advisor init and /career-advisor backfill first. Startup ideas built on documented expertise are 10x stronger than ideas from a blank page."
Key files to scan:
{CAREER_ROOT}/1-identity/accomplishments.md — problems you've solved, at what scale{CAREER_ROOT}/1-identity/skills-inventory.md — your unfair advantages{CAREER_ROOT}/1-identity/my-identity.md — values, ikigai, what energizes you{CAREER_ROOT}/2-target/dream-job.md — conditions you want (informs founder fit)
The Exploration Workflow
MINE EXPERTISE → SPOT PROBLEMS → PICK A NICHE → FRAME HYPOTHESIS → HAND OFF TO VALIDATION
Step 1: Expertise Mining
Scan the user's career data and extract:
| What to Find | Where to Look | Why It Matters |
|---|
| Repeated problems | accomplishments.md — what broke, what you fixed | Problems you've solved 3+ times are startup candidates |
| Expensive problems | Achievement results — $, time, risk quantified | Expensive problems = willingness to pay |
| Domain depth | skills-inventory.md — hard skills, years, proficiency | Your unfair advantage over generic founders |
| Network access | Past roles, industries, company sizes | Who you can reach for customer discovery |
| Energy patterns | my-identity.md — ikigai, values, archetypes | Founders quit when the problem bores them |
Output: A ranked list of Problem-Expertise Pairs — problems you understand deeply AND have the skills to solve.
Step 2: Problem Expansion
For each top Problem-Expertise Pair, expand into adjacent pain:
The "Who Else" Question Chain:
- "You solved this at [Company]. Who else has this problem?"
- "What size company hits this problem? At what stage?"
- "What do they do today? (DIY, hire someone, use a tool, ignore it)"
- "What does ignoring it cost them per month/quarter/year?"
- "How would you find 10 of these people this week?"
The "Why Now" Filter:
- What changed in the market that makes this problem solvable now?
- New technology? Regulation? Behavior shift? Platform change?
- If you can't answer "why now" — the idea may be valid but poorly timed.
Step 3: Niche Selection
Narrow from "this is a problem" to "this is MY customer":
The Niche Scorecard:
| Criterion | Score (1-5) | Notes |
|---|
| Problem severity | | How painful is this? Hair-on-fire (5) vs. mild annoyance (1) |
| Willingness to pay | | Do they already spend money on this? Budget exists? |
| Your access | | Can you reach 10 prospects this week without cold outreach? |
| Your expertise | | Could you solve this faster/better than 95% of people? |
| Market size | | Enough customers to build a real business? (>$1M TAM for solo) |
| Founder fit | | Does this align with your ikigai? Would you work on it for 3 years? |
Scoring rules:
- Any criterion < 2 = disqualify
- Total < 18 = weak niche, explore more
- Total 18-24 = promising, worth validating
- Total 25+ = strong candidate, move to hypothesis
The Beachhead:
Pick the narrowest possible first customer: "[Job title] at [company stage/size] in [industry] who [specific situation]."
Example: "DevOps leads at Series A startups with 5-20 engineers who just got their first $10K cloud bill."
Step 4: Hypothesis Framing
Convert the niche into a testable hypothesis using this template:
# Startup Hypothesis: [Working Name]
## The Problem
[Specific audience] struggles with [specific problem] because [root cause].
Today they [current solution], which costs them [quantified pain].
## The Insight
Because of my experience [doing X at Y], I know that [non-obvious insight
about the problem that outsiders miss].
## The Solution (Directional — NOT a spec)
A [product type] that [key capability], enabling [audience] to [outcome].
## Why Me
- [Relevant accomplishment from vault]
- [Unique skill or access]
- [Network advantage]
## Why Now
[Market change that creates the opening]
## Beachhead Customer
[Job title] at [company type] who [trigger event].
I can reach 10 of them by [specific channel].
## Key Assumptions to Test
1. [ ] [Assumption about the problem] — Test by: [method]
2. [ ] [Assumption about willingness to pay] — Test by: [method]
3. [ ] [Assumption about the solution] — Test by: [method]
## Next Step
→ Run `/idea-validation` with this hypothesis
Save to {CAREER_ROOT}/3-research/startup-ideas/ or a user-specified location.
Save-to-Disk Rules
Nothing stays in conversation only. Every exploration session must produce saved files:
| Output | Location | Format |
|---|
| Problem-Expertise Pairs | {CAREER_ROOT}/3-research/startup-ideas/explorations.md | Ranked list with scores and notes |
| Niche Scorecard | Append to explorations.md | Scorecard table per niche evaluated |
| Hypothesis | {CAREER_ROOT}/3-research/startup-ideas/HYPO-NNN-name.md | One file per hypothesis using template above |
| Pivot notes | Append to existing HYPO-NNN file | New section with what changed and why |
Create 3-research/startup-ideas/ on first use if it doesn't exist.
After saving, prompt: "Review what I wrote — I may have made the opportunity sound bigger than the evidence supports. Check that every claim maps to a real accomplishment or a real conversation."
Exploration Modes
| Command | What It Does |
|---|
/startup-explorer mine | Scan career data, produce ranked Problem-Expertise Pairs |
/startup-explorer expand [problem] | Run "Who Else" questions + "Why Now" filter on a specific problem |
/startup-explorer niche [problem] | Score a problem with the Niche Scorecard, define beachhead |
/startup-explorer hypothesis [niche] | Frame the testable hypothesis document |
/startup-explorer full | Run all 4 steps end-to-end |
/startup-explorer compare | Side-by-side scorecard of multiple niche candidates |
/startup-explorer pivot [hypothesis] | Rework an existing hypothesis after validation feedback |
Integration with Other Skills
| When | Skill | What Happens |
|---|
| Before this skill | career-advisor | Backfill roles, document accomplishments, build skills inventory |
| Parallel path | landing-page-service-discovery | Consulting-first path — package expertise as a service instead of (or before) building a product |
| After hypothesis | idea-validation | Test assumptions with Mom Test interviews, GO/PIVOT/KILL |
| After validation | business-model | Select revenue model, pricing, unit economics |
| After business model | fundraising | Build pitch deck if raising capital |
| After business model | go-to-market | Launch strategy if bootstrapping |
career-advisor ──→ startup-explorer ──→ idea-validation ──→ business-model ──→ fundraising
│ │
└──→ landing-page-service-discovery (consulting path) go-to-market
The Consulting-First Path
Many solo founders should start consulting before building a product. If the user's niche score is high on access and expertise but unclear on market size, recommend:
- Run
/landing-page-service-discovery to package expertise as a service
- Serve 5-10 clients manually — this IS customer discovery
- Notice the patterns: what every client asks for, what you automate, what scales
- THEN frame the startup hypothesis from real client pain, not imagination
This is the "do things that don't scale" approach. The career-advisor vault captures the evidence.
Anti-Patterns
| Anti-Pattern | Why | Do This Instead |
|---|
| Starting with the solution | "I want to build an AI tool for X" — you skipped the problem | Start with problems you've witnessed, not tools you want to build |
| Niche too broad | "Small businesses" = nobody | Narrow until you can name 10 specific people |
| No unfair advantage | "Anyone could build this" = commoditized on day one | Your startup should leverage YOUR specific expertise + network |
| Skipping "Why Now" | A good idea at the wrong time is a failed startup | If nothing changed recently, the timing may be wrong |
| Analysis paralysis | Exploring 20 ideas forever, validating none | Score 3-5 with the Niche Scorecard, pick the highest, go validate |
| Ignoring founder fit | High-scoring niche you don't care about = burnout in 6 months | Your ikigai matters — check it against my-identity.md |
| Building before talking | "Let me code the MVP first" — classic founder mistake | Talk to 10 potential customers before writing any code |
Power Move
"Scan my career folder and find the 3 most promising startup ideas hiding in my experience. Score each with the Niche Scorecard, recommend the strongest one, define the beachhead customer, and write a testable hypothesis I can take into /idea-validation today."
