---

name: client-report description: Generate client-friendly AI visibility reports from audit data. Use when creating client reports, visibility reports, or presenting audit results in a client-facing format. Triggers on "client report", "generate report", "visibility report", "create report", "report for client". allowed-tools: Read, Write, Edit, Glob, Grep, Bash

Client Report Skill

Generate client-friendly AI visibility reports that translate technical audit findings into compelling narratives clients can understand.

Core Principle

"Here's what we found -> Here's what we fixed -> Here's the proof"

Every report follows this narrative arc. No jargon. No technical terms. Just results.

The 8-Section Narrative Structure

Reports use these components (from aeo-landing/src/components/report/):

#	Component	Purpose
1	ReportCover	Brand name, category, audit date, tagline
2	HeadlineResult	3 most impactful before/after stats
3	CurrentVisibility	Scores per engine (branded vs discovery)
4	KeyFindings	What we discovered (synthesized, not raw)
5	Achievements	Gap analysis showing before/after improvements
6	ResultsProof	Recommendations with completion status
7	CompetitorPosition	Where client sits vs competitors
8	NextSteps	Pending actions, prioritized

Supporting components:

• SectionNav - Sticky navigation between sections
• SectionHeading - Consistent section headers
• ProgressRing - Visual score indicators

No-Jargon Rules

NEVER show these terms to clients:

Internal Term	Client-Facing Label
LLM	AI platforms
AEO	AI visibility
SERP	Google search results
Discovery queries	When people search for what you do
Branded queries	When people search your name
Grounding	How AI finds information
Consistency score	Recommendation rate
10-run test	We tested 10 times
Cache forcing	Making AI remember you
Schema.org	Structured data so AI reads your site correctly
SSR	Making your site readable by AI
robots.txt	AI crawler access
First 50 words	Opening content
Triangulation	Multiple sources confirming you

Data Structure Requirements

The report data file (src/data/[client]-report.ts) must export these TypeScript interfaces:

export interface VisibilityScore {
  engine: string;
  branded: number;    // 0-100
  discovery: number;  // 0-100
}

export interface Competitor {
  name: string;
  location: string;
  specialization: string;
  priceRange: string;
  tier: 1 | 2;
}

export interface GapItem {
  area: string;            // Internal technical area
  clientLabel: string;     // REQUIRED: Plain English label shown to client
  description: string;     // What was tested
  beforeScore: number;     // 0-100
  afterScore: number;      // 0-100
  status: "done" | "pending" | "in-progress";
  detail: string;          // Plain English explanation
}

export interface Recommendation {
  action: string;          // Internal action description
  clientLabel: string;     // REQUIRED: Client-facing action description
  priority: "immediate" | "short-term" | "medium-term";
  status: "done" | "pending" | "in-progress";
  category?: string;
}

export interface KeyFinding {
  title: string;           // Plain English headline
  detail: string;          // 1-2 sentence explanation
  severity: "high" | "medium" | "low";
}

Additionally export:

• brandInfo - Name, website, location, category, audit date, tagline, key stats
• executiveSummary - headline, subheadline, heroStat (averaged across engines for the key query)
• headlineResults - Array of 3 most impactful before/after changes
• visibilityScores - Per-engine branded/discovery scores
• engineDetails - Per-engine narrative detail + strength label
• gapAnalysis - Array of GapItems
• recommendations - Array of Recommendations
• competitors - Array of Competitors
• keyFindings - Array of KeyFindings

Key Principles

1. Every GapItem gets a clientLabel - Plain English, no jargon
2. Every Recommendation gets a clientLabel - Describes what was done in client terms
3. keyFindings are synthesized - Not copied raw from audit; rewritten as insights
4. headlineResults pick the 3 most impactful - Best before/after stats
5. executiveSummary.heroStat averages across engines - Average the before/after scores from ChatGPT and Gemini for the key discovery query
6. Scores are percentages (0-100) - Not fractions like 7/10
7. detail fields use plain language - "You weren't mentioned" not "0% consistency score"

Component Patterns

The page component follows this pattern:

import { ReportCover } from "../components/report/ReportCover";
import { SectionNav } from "../components/report/SectionNav";
import { HeadlineResult } from "../components/report/HeadlineResult";
import { CurrentVisibility } from "../components/report/CurrentVisibility";
import { KeyFindings } from "../components/report/KeyFindings";
import { Achievements } from "../components/report/Achievements";
import { ResultsProof } from "../components/report/ResultsProof";
import { CompetitorPosition } from "../components/report/CompetitorPosition";
import { NextSteps } from "../components/report/NextSteps";

export function [Client]Report() {
  return (
    <main className="min-h-screen max-w-5xl mx-auto px-4 pb-20">
      <ReportCover />
      <SectionNav />
      <HeadlineResult />
      <CurrentVisibility />
      <KeyFindings />
      <Achievements />
      <ResultsProof />
      <CompetitorPosition />
      <NextSteps />
    </main>
  );
}

Report Generation Workflow

1. Read the client's audit file: clients/[client]/[client]-aeo-audit.md
2. Read the client's playbook: clients/[client]/[client]-aeo-playbook.md
3. Extract brand info, scores, gaps, recommendations, competitors
4. Translate all findings into client-friendly language
5. Generate the data file: aeo-landing/src/data/[client]-report.ts
6. Create the page component: aeo-landing/src/pages/[Client]Report.tsx
7. Register the route in the router (if one exists)
8. Run build to verify: cd aeo-landing && npm run build

Verification Checklist

Before delivering a report, verify:

• Every GapItem has a non-empty clientLabel
• Every Recommendation has a non-empty clientLabel
• No jargon terms appear in any client-facing strings
• headlineResults has exactly 3 entries
• heroStat averages scores across engines for the key discovery query
• All scores are 0-100 percentages
• keyFindings are synthesized insights, not raw audit data
• The page component renders all 8 sections
• Build passes without errors

Reference

• Data structure example: aeo-landing/src/data/fuegenix-report.ts
• Page component example: aeo-landing/src/pages/FuegenixReport.tsx
• Report components: aeo-landing/src/components/report/