customer-panel-of-experts

Simulates a debate among data-grounded buyer personas to evaluate business decisions.

1|Updated Aug 8, 2026
One-click install
npx skills add https://github.com/th-efool/SKILLS --skill customer-panel-of-experts-th-efool
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: customer-panel-of-experts
Source: https://github.com/th-efool/SKILLS/tree/main/customer-panel-of-experts
Command: npx skills add https://github.com/th-efool/SKILLS --skill customer-panel-of-experts-th-efool

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? High-stakes decisions like price increases, product launches, and feature cuts are often made by guessing how customers will react. This Skill puts your actual buyer personas in the room, running a structured debate so you see objections, segment-level fallout, and a clear recommendation before you commit. ## Core Features & Use Cases - Persona-grounded debate: Loads a persona library (from icp-deep-scanner or connected data sources) and seats 3-6 relevant personas who argue in character from their real goals and pains. - Structured decision synthesis: Returns a GO / GO WITH CHANGES / NO / TEST FIRST recommendation, a per-persona vote table, ranked objections with blast radius, and the cheapest experiment to de-risk the biggest unknown. - Honesty guardrails: Read-only data connections, no real customer PII in output, and provisional panels are loudly labeled when not grounded in customer data. - Use Case: Before raising prices 20%, run the panel to learn which segments churn, which accept it, what objection will dominate, and what concession flips a NO to a YES. ## Quick Start Run the customer panel on this decision: should we raise our Pro plan price by 20% next quarter?

Frequently Asked Questions about customer-panel-of-experts

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I test a pricing increase with customer personas before launching it?▼

Frame the decision with the specific price change and success metric, then seat 3-6 personas including the economic buyer and a price-sensitive segment. The panel debates in character and returns a GO / GO WITH CHANGES / NO / TEST FIRST verdict with ranked objections and mitigations.

What data do I need to build realistic buyer personas for a decision panel?▼

The preferred input is a persona library and icp-profile.md produced by icp-deep-scanner from your connected tools. If none exists, the skill can generate one from connected data sources, or bootstrap 3-5 provisional personas from your description, clearly labeled as not data-grounded.

Can I use a persona debate panel without connecting my customer data tools?▼

Yes, but the panel falls back to provisional personas built from what you tell it, and the entire session is labeled PROVISIONAL. Grounded personas from real data are strongly preferred because guessed panels can masquerade as researched ones.

Does the customer panel expose real customer names or emails in its output?▼

No. Personas are archetypes, and the skill explicitly forbids surfacing real customer names, emails, or account IDs in the debate. Quotes are scrubbed, and connected tools are accessed read-only with secrets kept in environment variables.

When should I not rely on a simulated customer panel for a decision?▼

Avoid relying on it when the persona library is thin or missing a critical viewpoint, since the panel can only argue from the personas it has. The output includes a confidence and blind spots section, and cheap real-world experiments are recommended before betting the company.