Validate Your Business Idea with The Founder Claude Skills

What The Founder does

The Founder is a collection of eleven Claude skills designed to test a business idea before you invest heavily in launching it. The skills cover strategy, competition, customers, pricing, offers, finance, marketing, branding, operations, launch planning, and business-plan creation.

The workflow is intended to replace disconnected early-stage guesses with a structured investigation. Claude writes its work to a founder/ folder in your project, allowing later skills to read and build on earlier results.

The project is free, MIT licensed, requires no signup or API key, and does not require anything to be connected. Its included Python tools use the standard library and require Python 3.8 or newer.

The eleven skills

• /founder-board: Uses three strategic lenses to score an idea, identify risks, and vote to fund, fund if conditions are met, or pass.
• /founder-competitors: Researches direct competitors, indirect competitors, substitutes, prices, positioning, and customer complaints using linked public sources.
• /founder-consumer: Creates a simulated panel of buyer agents with different incomes, habits, and objections, then tallies who would buy and why.
• /founder-pricing: Analyzes the panel's price answers, compares them with competitors and margins, and suggests what to test.
• /founder-offer: Builds an offer stack with elements such as bonuses, a guarantee, urgency, and a name, then re-tests it against buyer objections.
• /founder-cfo: Calculates contribution per sale, profit margin, break-even volume, monthly year-one results, cash requirements, payback, and what-if scenarios.
• /founder-marketing: Develops positioning, channels, a dated 30-day campaign, ten hooks, and the maximum customer acquisition cost you can pay.
• /founder-brand: Produces name candidates and a brand brief, including trademark, domain, and social-handle checks to run.
• /founder-ops: Maps suppliers, staffing, day-one routines, tools, permits to check, and operational risks while feeding real costs back to the CFO.
• /founder-launch: Plans a low-cost real-world test, a dated countdown, launch-day activities, and the first 30 days.
• /founder-plan: Compiles the work into a business plan and computes a verdict of Profitable, Not yet, or Incomplete.

Install The Founder

Install through Claude

The simplest option is to paste the repository URL into Claude and ask it to install the skill:

https://github.com/Jakeschincariol/founder-skill

install skill

Install as a Claude Code plugin

In Claude Code, add the marketplace and install the plugin:

/plugin marketplace add Jakeschincariol/founder-skill
/plugin install founder-skill@founder-skill

Plugin installation namespaces the skills. You will use names such as /founder-skill:founder-board.

Copy the skills for plain commands

If you prefer commands such as /founder-board without the plugin namespace, clone the repository and copy the skill folders into your global Claude skills directory:

git clone https://github.com/Jakeschincariol/founder-skill.git
cp -r founder-skill/skills/founder-* ~/.claude/skills/

For project-local skills, copy the same folders into your repository's .claude/skills/ directory instead. If you are not using Claude Code, paste an individual SKILL.md file at the top of a Claude chat to run it as a mode. This standalone approach does not include the sub-agents or Python tools, but the method still works.

Run the skills in sequence

For a complete business assessment, run the skills in this order:

board -> competitors -> consumer -> pricing -> offer -> cfo -> marketing -> brand -> ops -> launch -> plan

Each skill reads the work produced by the previous skills from the founder/ folder. You can also run an individual skill when you need a focused answer. For example, /founder-consumer --quick is useful when you want an early signal about an idea you are still considering.

Work through an example idea

The README uses a matcha café in Toronto as an example, offering iced matcha lattes for $6.50. The same workflow applies to other products and services as long as you provide the relevant idea, customer, pricing, and operating assumptions.

1. Establish the idea and strategic risks

Start with /founder-board. It creates founder/idea.md from your description and presents the idea to three board members. Each board member applies a different published framework: the value of an offer, the possibility of building a monopoly, and product-focused thinking. The results include scores, risks, votes, and conditions that become a checklist for later work.

The board applies summaries of ideas associated with Alex Hormozi's $100M Offers, Peter Thiel and Blake Masters' Zero to One, and Walter Isaacson's biography of Steve Jobs. It does not speak as those people or claim their endorsement.

2. Research competitors

Run /founder-competitors next. For the café example, it maps nearby matcha bars and coffee shops, records latte prices, and identifies complaints in reviews. Research is based on public sources, and facts should be linked rather than invented.

3. Simulate buyer reactions

Use /founder-consumer to describe your target customer. For the matcha café, the example targets urban professionals aged 24 to 38 who buy coffee most days.

The panel creates buyer cards with different incomes, habits, and objections, then runs one sub-agent per buyer. The output identifies who would buy, who would pass, the reasons in their own words, and what might change a negative decision.

4. Set a price to test

/founder-pricing reads four price answers from each buyer and turns them into an acceptable price range. It also compares the proposed price with competitors and your margin requirements.

The included Van Westendorp tool calculates the points of marginal cheapness and expensiveness, along with the optimal and indifference prices. This lets you use either panel responses or price answers collected through a real survey.

5. Improve the offer

Run /founder-offer to address the most important objections. For example, the café could test a launch-week price or a loyalty card. The skill then re-tests the revised offer on a smaller panel to see whether the estimated buy rate changes.

6. Check the economics

/founder-cfo turns your assumptions into a financial model. The example reports $4.76 in contribution per cup, a break-even point of 186 cups per day, a 38% planned profit margin, $93,926 in year-one operating profit after a slow ramp, and payback for a $25,280 startup spend in month eight. It also identifies $34,092 in cash needed before the business pays for itself.

These figures are specific to the example inputs, not promises about another business. The tool traces every result to an input, treats estimates as estimates, and reports an error when a required field is missing instead of guessing.

7. Build the go-to-market plan

Use /founder-marketing, /founder-brand, /founder-ops, and /founder-launch to turn the research into execution. Together they cover a 30-day marketing campaign, a name and visual direction, suppliers and operating routines, and a dated launch plan that begins with a low-cost real-world test such as a pop-up.

8. Compile the final plan

Finish with /founder-plan. It assembles the material into founder/business-plan.md and calculates its verdict from the numbers. The result is Profitable only when every unit earns money, the first year is profitable, break-even fits the business's capacity, and enough of the consumer panel buys. Otherwise, the result is Not yet or Incomplete.

Use the included Python tools

The repository includes four standard-library Python tools. Nothing needs to be installed with pip.

python3 skills/founder-cfo/unit_economics.py founder/numbers.json --out founder/cfo.md
python3 skills/founder-consumer/panel.py init --customer founder/customer.json --pitch founder/pitch.md
python3 skills/founder-pricing/van_westendorp.py founder/panel/answers --price 6.50
python3 skills/founder-plan/compile.py --dir founder

Calculate unit economics

Pass a JSON file containing your business assumptions to unit_economics.py. It reports contribution, break-even, planned margin, monthly year-one results, cash required, payback, and three what-if scenarios. The repository includes an example input at skills/founder-cfo/example.json.

Generate a consumer panel

Use panel.py init with a customer JSON file and a pitch file. The panel deals persona cards using the same seed and writes one brief per buyer, validates the answers, and produces a tally. It does not answer for a buyer.

Analyze survey prices

Give van_westendorp.py the panel answer directory and a proposed price. The tool can analyze panel results or a CSV generated from a real survey.

Compile the plan

Run compile.py --dir founder after the other skills have written their outputs. It assembles the available material and computes the final verdict.

Advanced usage and practical tips

• Keep the workflow's files together. The skills depend on shared files in founder/, so use one project directory for the idea and its outputs.
• Run focused checks early. Use /founder-consumer --quick when you need an inexpensive first look before completing the entire sequence.
• Replace assumptions with evidence. Give the CFO supplier quotes and other real costs where possible, then confirm payroll, tax, and business structure with an accountant.
• Validate simulated demand in the real world. The consumer panel is made of simulated buyers, not customers. Treat its buy rate as an upper bound and use the launch test to confirm important decisions with real people.
• Respect research sources. Competitor research should use public sources, include links, and avoid scraping in ways that violate a site's terms.
• Run the tests. From the repository root, execute python3 -m unittest discover -s tests -v to run the included tests for the tools.

Important limitations

The consumer panel is simulated research, not a forecast or a source of testimonials. Language models can be agreeable, so validate major decisions with real customers.

The financial outputs are only as reliable as the inputs. The project is not financial, legal, or tax advice. Permits, trademarks, regulatory requirements, and claims depend on your location and should be confirmed with qualified professionals.

Conclusion

The Founder provides a structured way to pressure-test an idea before spending heavily on it. Install the skills in Claude, start with the board and competitor research, move through buyers, pricing, offers, and unit economics, then finish with launch execution and a computed business-plan verdict. Use its simulations and estimates as decision support, and confirm the most important assumptions with real people, suppliers, and professional advisers.

The project is available under the MIT license at The Founder GitHub repository.