case-study-writing

Overview

What is the case-study-writing skill?

The case-study-writing skill is a guided workflow for creating clear, credible B2B case studies using the STAR framework (Situation → Task → Action → Result). It is built as an inference.sh skill so your agent can research, outline, and draft customer stories that highlight real business outcomes.

Instead of starting from a blank page, this skill walks the agent through:

• Setting the Situation and customer context
• Defining the Task or challenge in concrete terms
• Describing the Action (your solution and implementation)
• Quantifying the Result with metrics, before/after comparisons, and visuals

Who is it for?

Use the case-study-writing skill if you need to produce:

• Customer success stories for SaaS, services, or B2B products
• Sales enablement collateral that showcases proof of results
• Marketing case studies for websites, decks, or campaigns
• Portfolio pieces for agencies, consultants, or freelancers

It fits teams in:

• Marketing and content marketing
• Customer marketing and customer success
• Sales and sales engineering
• Agencies and B2B consultants

What problems does it solve?

This skill is designed to solve common case-study problems:

• Stories that feel like fluff with no clear business impact
• Unstructured drafts that bury the most important results
• Weak or missing metrics that undermine credibility
• Lack of research into the customer’s industry and context

By following a fixed STAR structure and word-count guidance from the repository, the agent keeps each section focused and ensures the total length stays in the 800–1200 word range, balancing detail with readability.

When is case-study-writing a good fit?

Choose this skill when:

• You have a product or service with measurable outcomes
• You need repeatable, on-brand case study structure across multiple stories
• You want your agent to use research and data instead of generic marketing language
• You plan to reuse case studies in sales decks, landing pages, or social campaigns

It is not a good fit when:

• You only need short testimonials or one-line quotes
• You are writing purely technical documentation or user manuals
• You do not have permission to reference customer names, metrics, or results, and cannot anonymize them

How to Use

Prerequisites

The repository skill definition notes that it is designed to run with the inference.sh CLI (infsh).

You should have:

• Node.js installed (to use npx for skill install)
• Access to the infsh CLI if you plan to follow the Quick Start research step

If you intend to use the research flow shown in the source, install the inference.sh CLI following the official instructions at:

• https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md

Installation steps

Use the Agent Skills Finder install command to add the skill to your agent environment:

npx skills add https://github.com/inferen-sh/skills --skill case-study-writing

This pulls the case-study-writing skill definition and related guide files from the inferen-sh/skills repository under guides/writing/case-study-writing.

After installation, open:

• SKILL.md for the canonical description and workflow
• Any referenced guides under guides/writing/case-study-writing for detailed structure and examples

Core workflow with STAR

Once installed, you can prompt your agent to use case-study-writing whenever you mention triggers such as case study, customer story, success story, B2B case study, or client testimonial.

A typical workflow is:

1. Gather inputs
   Provide the agent with:

   • Customer name and industry (or an anonymized equivalent)
   • Product or service they used
   • The main challenge and any known metrics
   • Quotes or notes from interviews, if available

2. Research the customer’s context (optional, via infsh)
   The SKILL guide demonstrates using the inference.sh CLI for external research. For example:

   infsh login
   
   # Research the customer's industry
   infsh app run tavily/search-assistant --input '{
     "query": "SaaS customer onboarding challenges 2024 statistics"
   }'
   

   You can adapt this query to your customer’s industry (e.g., manufacturing, fintech, ecommerce) to collect current stats and trends that strengthen the Situation and Task sections.

3. Draft with the STAR structure
   The skill prescribes the following approximate structure and lengths:

   • Situation (100–150 words) – Who the customer is, their context, and why the status quo was not good enough.
   • Task (100–150 words) – The specific challenge or objective, described in business terms (e.g., reduce churn, improve lead quality, shorten onboarding).
   • Action (200–300 words) – What solution was implemented, how it worked, and key implementation steps that highlight your product or service.
   • Result (100–200 words) – Measurable outcomes with before/after comparisons, timelines, and key KPIs.

   In your prompt, tell the agent to:

   • Follow the STAR framework in order
   • Stay within the total 800–1200 word range
   • Clearly separate each section

4. Add metrics, visuals, and quotes
   The skill’s description emphasizes:

   • Metrics presentation – Have the agent call out metrics like conversion rate, revenue impact, time saved, or error reduction. Ask it to propose simple visual ideas (e.g., before/after bar chart, funnel drop-off chart) that a designer could later implement.
   • Customer quotes – If you provide raw interview notes, ask the agent to suggest 1–3 punchy quotes aligned to the Task, Action, and Result sections.

5. Plan distribution formats
   Once the main narrative is approved, you can prompt the agent (using the same skill context) to adapt the case study into:

   • A one-page sales PDF outline
   • Slide headlines for a sales deck
   • Short social media posts highlighting the key result
   • Website or blog case study summary copy

Adapting the skill to your workflow

The repository recommends treating the skill as a workflow template, not a rigid script. You can:

• Adjust section lengths if your industry demands more technical depth
• Swap in your own research tools or data sources instead of the example tavily/search-assistant
• Integrate the STAR outline into your CMS or proposal templates

When configuring your agent, map your preferred trigger phrases (e.g., “draft customer success story” or “write SaaS onboarding case study”) to case-study-writing so the workflow activates consistently for your team.

FAQ

What does the case-study-writing skill actually do?

The case-study-writing skill guides your agent through a structured B2B case study workflow using the STAR framework. It emphasizes research-driven context, clear problem definition, detailed explanation of your solution, and results backed by metrics, quotes, and suggested visuals.

How do I install case-study-writing?

Run:

npx skills add https://github.com/inferen-sh/skills --skill case-study-writing

Then open SKILL.md in the installed repository to see the full instructions, including the Quick Start and STAR framework table.

Do I have to use the inference.sh CLI (infsh)?

The SKILL guide shows example commands using the inference.sh CLI (infsh) to perform research with tavily/search-assistant. This is recommended if you want automated industry research, but you can also:

• Run similar research workflows with your own tools, or
• Provide the agent with your own pre-collected research instead.

The case-study-writing structure itself (STAR and word-count guidance) can still be followed without infsh.

What information should I prepare for the best results?

To get strong case studies from the case-study-writing skill, prepare:

• Customer background (industry, size, location, product usage)
• The main business challenge or objective
• Implementation details (how your product/service was used)
• Quantitative results (KPIs before and after)
• Any interview notes or customer quotes

The richer the inputs, the more specific and credible the final case study will be.

Can I use case-study-writing for non-B2B stories?

The skill is optimized for B2B scenarios with measurable business metrics. You can adapt it to B2C or internal change stories, but the structure, tone, and KPI focus were designed specifically for B2B case studies.

How long will the finished case study be?

According to the STAR framework guidance in the repository, the recommended length is 800–1200 words. This is usually enough to show context, explain your solution, and prove results without overwhelming readers.

Is this skill suitable for social media posts or ads?

case-study-writing is primarily a report-style, long-form workflow. It’s well suited for full case studies that you can then repurpose into:

• Social media snippets
• Sales outreach emails
• Slides or one-pagers

For pure ad copy or short-form social content, you may want to pair this skill with more specialized copywriting skills and use the case study as the source material.

Where can I see all files and rules for this skill?

After installation, open the skill directory in your environment to view:

• SKILL.md – primary description and usage guidance
• Any additional guides under guides/writing/case-study-writing

Use the Files tab in your Agent Skills Finder or your local editor to inspect the full tree, including any nested references or helper materials provided by the repository.