agentleFS
styfinitystyfinity/claude-marketing-department/skills/case-study-writer/SKILL.md

case-study-writer

Skill6 starsChanged 4 months agoMITView on GitHub

Turns a real customer result into a structured case study — situation, sticking point, intervention, numbers, quotes — plus a short social version cut from the same material. Use when a genuine result exists and needs to work as proof across the funnel.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 2

    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:

    Every number and quote traces to a real source.

    line 33
  2. Distinctive:

    Keep the customer the hero.

    line 36

“Common” is measured: kept by at least 6 in 10 of the 8 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 Writer — proof, structured

A case study is not a testimonial with padding. It is the story of a decision the reader is about to make, told by someone one step ahead of them. The structure does the selling: a before-state the reader recognises as their own, a mechanism they can believe, and numbers they could check.

Inputs

  • The customer result and all raw material — notes, metrics, quotes, timeline: $ARGUMENTS
  • From marketing-brief.md if present: the ICP (so the story is framed for the right reader), the language bank, positioning & message, the banned-phrases list, and any rules on naming customers publicly.

Do this

  1. Inventory the raw material: who the customer is (and what may be named publicly), the before-state, what was actually done, the numbers with their timeframe, and verbatim quotes. List what is missing and ask once — none of it gets invented.
  2. Pressure-test the numbers. Baseline, timeframe, and what else was happening that could explain the change. A number without a baseline is a claim, not proof — present it honestly or leave it out.
  3. Write the case study in this shape:
    • Title — result-led and specific, no hype adjectives.
    • The situation — the before-state in the customer's words; the target reader should say "that's me" by line three.
    • The sticking point — what they had already tried and why it failed.
    • The intervention — what was done, step by step. The mechanism is the persuasive part; show the work.
    • The numbers — baseline to result, timeframe, context.
    • In their words — real quotes only, lightly trimmed, never improved.
    • What this means for you — the bridge to the reader, one CTA.
  4. Cut the social version: hook, three-beat arc, one number, CTA — reshaped for the feed, not compressed from the long copy.
  5. Run the naming pass. Apply the brief's rules on naming customers: keep the numbers, drop names where required, and note anything that needs the customer's sign-off before publishing.

Output

The full case study, the social version, and a proof-integrity note: the source of every number and quote, plus what to confirm with the customer before it ships.

Rules

  • Every number and quote traces to a real source. Nothing composite, nothing illustrative — if it is not real, there is nothing to write.
  • Curiosity belongs in the title, never in the numbers.
  • The intervention sells the method, not the vendor. Readers trust work they can see.
  • Keep the customer the hero. The company is the guide who turned up with a map.

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.