← ClaudeAtlas

kai-case-studylisted

Produce customer case studies from interviews or data — Problem, Solution, Results structure with perception engineering and quality gates. Use when "case study", "customer story", "testimonial", "success story", "client results", or any request to document a customer win.
cgallic/kai-cmo-harness · ★ 47 · AI & Automation · score 73
Install: claude install-skill cgallic/kai-cmo-harness
# /kai-case-study — A Customer Win Told In Numbers The Customer Approved ## Objective A publishable customer case study built on a Problem / Solution / Results arc, led by the result rather than the logo, carrying real numbers and the customer's own language — plus the derived assets that make it usable: a sales one-pager, standalone pull quotes, and headline variants per channel. Permission is the gate everything else waits behind. Named use requires the customer's approval; without it, the study is anonymized or it does not ship. ## Done when Work type `blog-post` — floor **E5/C3/O3** (`harness/eco-floors.yaml`). - **E5** — the live page returns 200 and matches the approved body, verified by someone other than the writer. Approval from the customer for named use is recorded before publication, not after. - **C3** — Four U's ≥ **12/16** (`python scripts/quality_gates/four_us_score.py <file>`), zero Tier 1 banned words (`python scripts/quality_gates/banned_word_check.py <file>`), zero AI slop phrases, `seo_lint` clean where the study publishes as a search-facing page, and a named non-producer reads it end to end. Max 2 auto-retry cycles, each naming the specific failing dimension; after two, surface to a human with the failures listed. - **O3** — organic clicks, impressions, indexation, or assisted conversions read from Search Console at the declared window. Sales-deck usage counts as a second declared metric when the study's primary job is enablement. ## Constraints