What if the customer never replies with numbers?
The pack produces a qualitative version that still works, and it marks the missing metrics rather than estimating them. Unverified numbers on a case study are worse than none.
Marketing
Catch the moment a customer is happy and turn it into proof you can put on the page.
Follows the Agent Skills open format, works with Claude Code, Cursor, Hermes Agent, OpenClaw, and 30+ compatible agents.
What it does
This skill starts from something a customer already said or a result you already delivered, and turns it into social proof. It writes the testimonial request, the specific questions to ask, and once you have answers, a case-study draft with the numbers up front.
Most teams do not lack wins. They lack a habit of asking while the win is fresh. The pack makes the ask small and specific, which is what gets a reply: three questions, a sentence explaining where the quote will appear, and a pre-written version the customer can approve or edit.
The case-study draft follows a structure buyers actually read: the situation before, what changed, the measurable result, and a quote that sounds like a person rather than a marketing department. Anything you have not verified is flagged rather than smoothed over.
Who it is for
Requirements
Customer signal
A quote, support thread, review, or result worth building on
Account context
What the customer does, what they bought, and any numbers you already have
Where it will run
Optional. Homepage, sales deck, or case-study page shapes the length.
Sample output
Source: Support thread, Harbor Loop (logistics scheduling, 40 staff) Signal: 'We cut our Monday planning meeting from two hours to twenty minutes.' Request email (send within 48 hours): Subject: Can we quote you on the Monday meeting thing? Hi Priya, you mentioned the Monday planning meeting went from two hours to twenty minutes. Three quick questions if you have five minutes: 1. What did that meeting cost the team before? 2. What changed first? 3. What would you tell another ops lead considering this? Here is a version you can approve or rewrite: 'Our Monday planning used to eat two hours. It now takes twenty minutes, and nobody dreads it.' Case-study draft headline: Harbor Loop cut weekly planning from 2 hours to 20 minutes Before: Eight people, one recurring meeting, no shared agenda Change: Scheduling moved out of the meeting and into a weekly brief Result: 1 hour 40 minutes recovered per week across eight people Unverified: the eight-person headcount is from the sales call, confirm before publishing.
What's in this pack
Included: Outcome definition
What exact job this pack solves
Included: Operating instructions
Step-by-step workflow the agent follows
Included: Input contract
Required and optional fields defined as a schema
Included: Output contract
Exact output structure the agent must return
Included: Worked examples(6 files)
3 standard, 1 ambiguous, 1 bad-fit, and 1 edge-case scenario
Test & eval set
FAQ
The pack produces a qualitative version that still works, and it marks the missing metrics rather than estimating them. Unverified numbers on a case study are worse than none.
It can draft from your own records, but every claim sourced only from your side is flagged. Anything customer-attributed needs their approval before you publish it.
Within 48 hours of the win. Reply rates drop sharply after that, which is the main reason most teams never collect the proof they have earned.
Bundle option
This pack is also included in OutcomeKit Complete, which groups related workflows for a faster setup.
View bundleRelated packs
Marketing
Get conversion-focused marketing copy with headlines, body text, CTAs, and social proof placement.
Test cases with expected outputs and pass/fail criteria
Included: QA checklist
Human review checklist for output quality
Marketing
Turn a product update into launch posts for X, LinkedIn, and email.
Sales
Score inbound leads for fit, urgency, and next step so reps or founders know where to spend attention first.