agentleFS
JakeschincariolJakeschincariol/whale-skill/case-study/SKILL.md

case-study

Skill15 starsChanged 6 days agoMITView on GitHub

Case Study. Turns the first customer into proof, with their permission. Once the product has shipped, it writes a short before-and-after from their real numbers and their real words, and keeps it marked as a draft until they have approved that exact text in writing. Change a word and it asks again. Use when the user says "case study", "testimonial", "customer story", "get a quote from them", "write up the result", "social proof", "can I use their name", or once a customer has seen results.

17 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use audit.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • OpenAISkill

    audit

    Audit or critique a product flow, journey, workflow, funnel, onboarding path, checkout path, settings path, screen, or multi-step product experience by capturing screenshots first, then reporting UX, design, and accessibility findings inline from that evidence. Use Figma only when the user explicitly asks for a board. Use when the user asks to audit, review, critique, inspect, assess, analyze, evaluate, or give feedback on a product experience.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 10 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 16 edited versions we read keep these.

  • 10

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before you have customers, you sell on the demo and the refund.

    line 14
  2. Distinctive:

    Get the real numbers from them: Before and after, and where they came from.

    line 32
  3. Distinctive:

    A real name and a real company,: if they agree. If they'd rather not be named, say "a regional freight company" and keep the number.

    line 42
  4. Distinctive:

    Typical, or not: If one customer's result is unusually good, don't suggest everyone gets it.

    line 45
  5. Distinctive:

    After North Pier Logistics has the product: Before, paid $6,400 a month to chase invoices.

    line 49
  6. Distinctive:

    Never writes a case study before delivery, and never makes up a quote, a number or a logo.

    line 55
  7. Distinctive:

    Never publishes or sends anything without a yes, and never uses a draft their approval doesn't cover.

    line 56
  8. Distinctive:

    Never presents a pre-sale payment as a result.

    line 57
  9. Distinctive:

    Stop at the first "no". Not wanting a case study never affects how the customer is treated.

    line 62
  10. Distinctive:

    Never invent a number, a result, a client or a quote.

    line 63

“Common” is measured: kept by at least 6 in 10 of the 16 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

case-study

Before you have customers, you sell on the demo and the refund. After you have one, you sell on them. One real story, with a name and a number, is worth more than anything you could write about yourself. But it's a claim about a real company, so it has to be true and it has to be theirs to give.

${CLAUDE_SKILL_DIR} below is this skill's folder (where this SKILL.md lives).

python3 "${CLAUDE_SKILL_DIR}/casestudy.py" draft --deal north-pier --before "Paid \$6,400 a month to chase invoices" \
    --after "Cancelled the collections service" --source "North Pier's own numbers, Jan 2027" \
    --quote "We stopped thinking about late invoices." --quote-by "Renee Alvarez, CFO"
python3 "${CLAUDE_SKILL_DIR}/casestudy.py" approve --deal north-pier --said "Approved, go ahead." --at 2027-02-10
python3 "${CLAUDE_SKILL_DIR}/casestudy.py" check --deal north-pier

Steps

  1. Wait for a result. draft refuses until the deal is delivered (/pre-sale deliver). Give it long enough to have a real number.
  2. Ask first. "Send this to Renee: would you be open to a short write-up, if you approve every word? Reply yes or no." A no is fine, and it ends here.
  3. Get the real numbers from them. Before and after, and where they came from. Ask what changed, in their words: "What's different now?"
  4. Draft. casestudy.py draft. It writes whale/case-studies/<deal>.md with a DRAFT line at the top. Keep it under 150 words. A quote is what they actually said or wrote, word for word, with their name. Don't tidy it into something they wouldn't say.
  5. Send them the exact text. "Send this to Renee for approval? Reply yes or no."
  6. Log their approval. casestudy.py approve --said "<their words>". It removes the DRAFT line, for this text only.
  7. Check before every use. casestudy.py check. If the text changed after they approved it, it says to ask again.
  8. Use it. On the next pitch, in /offer-builder, on the launch page. Only then does the "you'd be one of the first" answer in /objection-handler change.

What makes it believable

  • A real name and a real company, if they agree. If they'd rather not be named, say "a regional freight company" and keep the number.
  • One number. "Cancelled a $6,400-a-month service" beats a list of adjectives.
  • The bad part. What was hard before, or what took longer than expected. It's how people know it's true.
  • Typical, or not. If one customer's result is unusually good, don't suggest everyone gets it. The FTC treats testimonials as claims you must be able to back up.

Example from the reel's board

After North Pier Logistics has the product: Before, paid $6,400 a month to chase invoices. After, cancelled the collections service. Quote: "We stopped thinking about late invoices." Renee Alvarez, CFO. It stays a draft until Renee approves it. (Illustrative: this result and this quote are made up to show the format, and North Pier and Renee are fictional. Your case study uses only what your customer gives you.)

What it never does

  • Never writes a case study before delivery, and never makes up a quote, a number or a logo.
  • Never publishes or sends anything without a yes, and never uses a draft their approval doesn't cover.
  • Never presents a pre-sale payment as a result. "They paid" is not "it worked".

Consent and truth

  • Nothing is sent or published without the user's explicit yes for that exact message, and the customer's written approval of that exact text. One yes per message. Never "yes to all".
  • Stop at the first "no". Not wanting a case study never affects how the customer is treated.
  • Never invent a number, a result, a client or a quote. If they withdraw their approval, take it down.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.