菜单
Use this skill when the user needs to size a market, analyze competitors, calculate TAM/SAM/SOM, or validate a business idea. Covers market sizing, competitive analysis frameworks, napkin math, and bottom-up revenue estimation.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use this skill when the user wants to write long-form prose — blog posts, founder essays, build-in-public posts, About pages, newsletter intros, or origin stories. Also use when the user mentions 'blog post,' 'essay,' 'long-form,' 'founder voice,' 'build in public,' 'BIP post,' 'About page,' 'origin story,' or 'newsletter copy.' For microcopy and CTAs, see copywriting. For removing AI patterns from existing text, see humanize. For SEO articles, see seo-content. For content strategy and distribution, see content.
Use this skill when the user doesn't know what to work on next, asks 'what should I do?', wants to find high-value opportunities, or needs help deciding where to spend their time. Also use when the user mentions 'what should I do next,' 'what should I work on,' 'where should I focus,' 'find low-hanging fruit,' 'what's the most valuable thing I could do,' or 'I don't know what to do.' Scans the full project — code, infrastructure, marketing, growth, business — and surfaces 3-5 high-impact, low-effort opportunities ranked by value. Routes each recommendation to the right skill.
Use this skill to find and eliminate duplication across your codebase — UI components, database schema, and workflow logic. Also use when the codebase feels bloated, features take longer to build, changes break in unexpected places, or after significant AI-assisted development. Covers code deduplication, component reuse, database normalization, and modular architecture.
Use this skill when the user's app feels slow, the codebase feels bloated, or after significant development work. Also use when the user says 'my app is slow,' 'clean up my code,' 'reduce bundle size,' 'my hosting bill is too high,' or 'everything feels sluggish.' Optimizes across four dimensions: Speed (page load, API response), Code (unused files, dead code), Database (orphaned data, schema hygiene), and Dependencies (package bloat, bundle size).
Use this skill when the user is building software for a regulated industry and needs to understand compliance requirements. Also use when the user mentions 'HIPAA,' 'FERPA,' 'SOC 2,' 'PCI,' 'GDPR,' 'compliance,' 'regulated industry,' 'healthcare app,' 'fintech,' 'edtech,' or asks 'do I need to worry about [regulation]?' Covers what regulations mean for your tech stack — not legal theory, but concrete technical requirements.
Use this skill when the user needs to hire a developer, find a technical contractor, vet engineering talent, manage freelancers, write a job brief, decide between hiring and contracting, or bring on their first technical team member. Also use when the user says 'I need a developer,' 'should I hire someone,' 'how do I find a contractor,' 'I need technical help I can't do myself,' or 'AI keeps getting this wrong, I need a human.' Covers contractor sourcing, vetting, management, and the founder-developer working relationship for non-technical founders.
| name | market-research |
| description | Use this skill when the user needs to size a market, analyze competitors, calculate TAM/SAM/SOM, or validate a business idea. Covers market sizing, competitive analysis frameworks, napkin math, and bottom-up revenue estimation. |
Markets matter more than ideas. This skill helps you evaluate whether your market is worth entering — and how to validate demand before you build.
Before doing market sizing, validate that real people have this problem.
Method 1: Cold Outreach (Fastest — 1-2 weeks)
Cold outreach template:
Subject: Quick question about [problem area]
Hi [Name],
I'm researching how [ICP role] handles [specific problem]. I noticed you [signal that they have this problem — e.g., posted about it, work in relevant role, use a competing tool].
Would you be open to a 15-minute call this week? I'm not selling anything — just trying to understand the problem better before I build a solution.
Happy to share what I learn from the research as a thank you.
[Your name]
Method 2: Communities (Higher quality — 2-4 weeks)
Method 3: Landing Page Test (Lowest effort — 2-3 weeks)
Decision threshold:
5 conversations reveal 80% of what you need to know. The other 20% you learn by building and shipping. Don't wait for certainty — it never comes.
Tell AI:
Help me size the market for my SaaS idea:
- Product: [what it does]
- Target customer: [who — job title, company size, industry]
- Problem it solves: [the pain point]
Do a bottom-up calculation:
1. Estimate how many potential buyers exist (use industry data, LinkedIn job title counts)
2. Multiply by a realistic price point for this type of tool
3. Apply a 1-3% capture rate for a new entrant
4. Tell me: does this math support a solo-founder business ($200K-500K ARR)?
1. How many potential buyers exist?
[Industry size] × [% that match your ICP]
Example: 500,000 US marketing agencies × 12% that are 5-20 person = 60,000
2. What will each pay annually?
[Price point] × [12 months]
Example: $49/month × 12 = $588/year
3. Realistic addressable market:
[Buyers] × [Annual price] × [Capture rate]
Example: 60,000 × $588 × 2% = $705,600/year achievable
4. Sanity check: Does that number fund the business you want?
Solo founder needs $200K-$500K ARR to replace income + reinvest.
If max realistic capture is $100K, the market is too small.
TAM = Total market revenue (all possible buyers globally)
SAM = Segment you can serve (geography, vertical, size)
SOM = What you can realistically capture in 3 years
Rule of thumb: SOM is 1-5% of SAM for a new entrant.
If SOM doesn't fund your business, stop here.
Tell AI:
Analyze the competitive landscape for my product:
- My product: [what it does]
- My target customer: [who]
- Known competitors: [list any you know, or "not sure"]
For each competitor, find: pricing, target audience, strengths, weaknesses
(check G2/Capterra reviews for complaints). Then build a 2x2 positioning
matrix showing where I could differentiate.
Every product competes with four types:
For each direct competitor:
## [Competitor Name]
**What they do**: One sentence.
**Target audience**: Who they sell to.
**Pricing**: Tiers and price points.
**Strengths**: 2-3 things they do well.
**Weaknesses**: 2-3 gaps, complaints, or underserved areas.
**Positioning**: Homepage headline — how they describe themselves.
**Traffic/Scale**: Monthly visits (SimilarWeb), review count (G2/Capterra).
### Where You Win
What specific thing will you do 10x better?
Build a 2×2 matrix with the two dimensions your ICP cares most about:
Example axes:
X: Simple ←————→ Powerful
Y: Cheap ←————→ Premium
Plot competitors on this grid. Find the empty quadrant. That's your positioning opportunity.
Score each criterion 1-5. Minimum viable total: 25/40.
| Criterion | Score | Notes |
|---|---|---|
| Pain intensity — Hair-on-fire problem? | /5 | 5 = actively searching for a fix |
| Willingness to pay — Already pay for substitutes? | /5 | 5 = pay $50+/mo for something worse |
| Reachable audience — Can you find and contact them? | /5 | 5 = concentrated in known communities |
| Fragmented competition — No dominant monopoly? | /5 | 5 = many small players, no clear winner |
| Small enough for solo — Can one person serve this? | /5 | 5 = low support burden, self-serve viable |
| Recurring need — Will they use it monthly? | /5 | 5 = daily/weekly active use |
| Your unfair advantage — Domain expertise or distribution? | /5 | 5 = deep insider knowledge |
| Technical feasibility — Build core in 4-8 weeks? | /5 | 5 = well-understood, no R&D needed |
Scoring:
| Channel | Cost to Test | Time to Signal | Scalable? |
|---|---|---|---|
| Cold outreach | $0 (time only) | 2-4 weeks | Somewhat |
| Communities | $0 (time only) | 1-3 months | Somewhat |
| Paid search | $200-500 | 1-2 weeks | Yes (if CAC works) |
| SEO / Content | $0 (time only) | 3-6 months | Yes |
| Directories | $0-200 | 1-2 months | Yes (passive) |
| Partnerships | $0 (time only) | 2-6 months | Yes |
Rule: Pick the ONE channel you can test in <30 days with <$500. Prove it works before diversifying.
Confirm your ICP can actually buy:
All five should be "yes" for a solo-founder SaaS.
| Mistake | Fix |
|---|---|
| Analysis paralysis — researching for months instead of shipping | 5 conversations + napkin math. Then build. |
| Inflating TAM to feel good ("it's a $10B market!") | Bottom-up only. TAM is a vanity metric for solo founders. |
| Ignoring "do nothing" as a competitor | Most of your market tolerates the pain. Your real competition is inertia. |
| Researching the wrong ICP | Talk to the person who writes the check, not the person who has the problem. |
| Only talking to people who confirm your idea | Seek disconfirming evidence. Ask "why wouldn't you use this?" |
| Skipping competitive pricing research | Visit every competitor's pricing page. Screenshot them. This is your anchor. |