agentleFS
jpoindexterjpoindexter/design-case-study-skills/skills/case-study/SKILL.md

case-study

Skill1 starsChanged 4 months agoMITView on GitHub

Route and produce evidence-backed product or design case studies from raw notes, research, screenshots, specifications, decision logs, architecture, metrics, and project artifacts. Use when the user invokes /case-study, asks to write, critique, structure, or polish a portfolio case study, needs to explain a complex design project, or needs an AI-agent case study covering autonomy, human oversight, telemetry, trust, latency, failures, and recovery.

19 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

4 rules. 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 18 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Never convert a proposal into an outcome.

    line 49
  2. Distinctive:

    Run the accuracy, clarity, and brevity edit.

    line 60
  3. Very specific:

    Run the quality gate in references/quality-gate.md.

    line 62
  4. Very specific:

    Use assets/case-study-template.md when creating a new Markdown artifact.

    line 65

“Common” is measured: kept by at least 6 in 10 of the 18 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 Dispatcher

Resolve the project, intended reader, available evidence, and requested format. Mark missing facts; never fill them with plausible fiction.

Route

NeedLoad
Flat timeline or missing storynarrative-arc-builder
Screens without rationaleextract-design-rationale
Architecture shapes the experiencesystem-architecture-translator
Imprecise or inflated technical copytechnical-microcopy-editor
Dense, repetitive, slow-to-scan draftenforce-scannable-density
AI agent or probabilistic systemai-agent-case-study plus the applicable skills above
Complex multichannel or branching journeycase-study-storytelling and case-study-story-map
Deep stakeholder decision narrativecase-study-design-decisions
General long-form drafting and finishingcase-study-writing

Load the smallest set that covers the evidence and requested deliverable. When the user asks for a complete build from raw artifacts, apply the automated pipeline below as one workflow, not as separate reports.

Automated pipeline

Raw logs / notes / code / research
  → narrative-arc-builder
  → system-architecture-translator (when architecture affects the experience)
  → extract-design-rationale
  → technical-microcopy-editor
  → enforce-scannable-density
  → quality gate

For an AI-agent case study, apply ai-agent-case-study across every stage. It governs autonomy, permission, observability, uncertainty, intervention, and recovery; it is not an extra standalone report.

Evidence contract

Label every substantive claim:

  • Executed — supported by a test, trace, artifact, or measured output.
  • Observed — supported by research notes, usability observations, or recorded behavior.
  • Reported — attributed to a stakeholder, participant, or project record.
  • Proposed — a recommendation, hypothesis, future metric, or unverified narrative bridge.

Never convert a proposal into an outcome. Never invent research, metrics, quotes, responsibilities, constraints, alternatives, or impact. Use [evidence needed: …] where the story has a material gap.

Workflow

  1. Inventory source artifacts and create an evidence ledger.
  2. State the protagonist, goal, inciting problem, constraints, pivotal decisions, delivered change, and measured result.
  3. Select the smallest route above.
  4. Build the narrative arc before drafting sections.
  5. Translate consequential system behavior into visible states, controls, and recovery.
  6. Extract the decision rationale and rejected alternatives.
  7. For AI systems, document supervision and uncertainty before visual polish.
  8. Run the accuracy, clarity, and brevity edit.
  9. Enforce scannable density without deleting evidence.
  10. Run the quality gate in references/quality-gate.md.
  11. Deliver one clean case study plus a short unresolved-evidence list.

Use assets/case-study-template.md when creating a new Markdown artifact.

Done

A case study is ready only when a skeptical reader can answer:

  • What changed, for whom, and under what constraints?
  • What did the designer actually own?
  • Which decisions were difficult, what alternatives existed, and why were they chosen?
  • What evidence supports the outcome?
  • What remains unknown, limited, or confidential?
  • For AI agents: where does autonomy stop, how does a human intervene, what is observable, and how does failure recover?

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.