Imported from oyi77/1ai-skills (marketing/ai-seo/SKILL.md). Install upstream with npx skills add oyi77/1ai-skills --skill ai-seo. Copyright stays with the author (Apache-2.0).

Ai Seo

Overview

AI SEO (Answer Engine Optimization / GEO) optimizes content for AI-powered search engines — Perplexity, ChatGPT Search, Google AI Overviews, Gemini, Copilot — so your pages get cited in zero-click answers instead of buried below them. Unlike traditional SEO that chases ranking clicks, this skill restructures existing content into citation-ready formats, implements the schema AI crawlers prefer, and builds topical authority clusters that engines trust as primary sources.

When to Use

Trigger phrases:

• "ai seo"

• "Adapting SEO strategy for AI search"

• "Optimizing for Google AI Overviews"

• "Building content that AI engines cite"

• "Answer engine optimization (AEO)"

• "Generative engine optimization (GEO)"

• Adapting SEO strategy for AI search

• Optimizing for Google AI Overviews

• Building content that AI engines cite

• Implementing structured data markup

• Optimizing for Perplexity, ChatGPT Search, Gemini, Copilot

When NOT to Use

• When the audience is too small to justify the effort
• For regulated industries without compliance review
• When the campaign budget does not support the channel
• When the client has zero existing content to optimize

Money-Making Overview

Buyer persona: SME owners ($2M-$20M ARR), SaaS marketing directors, e-commerce brand managers, and agency clients who see organic traffic from Google dropping and want to capture the growing share of zero-click AI-generated answers in Perplexity, ChatGPT Search, and Google AI Overviews.

Problem they pay to solve: Traditional SEO targets clicks. AI search engines answer the question in the SERP — if your content isn't structured for extraction, you get zero visibility even when ranking #1. Clients lose 30-60% of referral traffic as users stop clicking through.

What you sell: AI answer engine optimization — restructuring existing content into citation-ready formats, implementing schema that AI crawlers prefer, and building topical authority clusters that AI engines cite as primary sources.

Pricing tiers:

First-dollar timeline: Day 1 — run audit script, deliver report by Day 3. Close first client at $500 within the first week by auditing 3 competitor sites for free, then selling the fix.

First Action in 60 Minutes

Run this script on your own or a target site to generate an instant AI search engine readiness report:

#!/usr/bin/env bash
# ai-seo-audit.sh — Usage: ./ai-seo-audit.sh https://example.com
set -euo pipefail

SITE="${1:?Usage: $0 https://example.com}"
echo "=== AI SEO Audit: $SITE ==="
echo ""

# 1. Check structured data (JSON-LD presence and types)
echo "--- Structured Data ---"
SD=$(curl -sL "$SITE" | grep -oP '(?<=<script type="application/ld\+json">).*?(?=</script>)' | head -5 || true)
if [ -z "$SD" ]; then
  echo "FAIL: No JSON-LD structured data found. AI engines rely on schema."
  echo "Fix: Add Organization, Article, or FAQPage schema."
else
  TYPES=$(echo "$SD" | grep -oP '"@type"\s*:\s*"[^"]+"' | sort -u)
  echo "Found schema types:"
  echo "$TYPES"
fi
echo ""

# 2. Check direct answer readability (H2 headings as question phrases)
echo "--- Answer Readiness ---"
H2S=$(curl -sL "$SITE" | grep -oP '<h2[^>]*>.*?</h2>' | sed 's/<[^>]*>//g' | head -10 || true)
QUESTION_COUNT=$(echo "$H2S" | grep -ciP '(how|what|why|when|where|which|can|does|is|are)\b' || true)
echo "H2s containing question phrasing: $QUESTION_COUNT"
if [ "$QUESTION_COUNT" -lt 3 ]; then
  echo "WARN: Few question-format headings. AI engines extract answers from Q&A patterns."
  echo "Fix: Convert H2 headings to natural-language questions your audience asks."
fi
echo ""

# 3. Check for definition/explanation paragraphs after headings
echo "--- Direct Answer Format ---"
CONTENT=$(curl -sL "$SITE" | sed 's/<[^>]*>/\n/g' | sed '/^$/d' | head -200 || true)
SHORT_PARAS=$(echo "$CONTENT" | awk 'length>0 && length<120' | wc -l)
TOTAL_PARAS=$(echo "$CONTENT" | awk 'length>0' | wc -l)
if [ "$TOTAL_PARAS" -gt 0 ]; then
  PCT=$((SHORT_PARAS * 100 / TOTAL_PARAS))
  echo "Short/direct paragraphs (under 120 chars): $PCT%"
  if [ "$PCT" -lt 20 ]; then
    echo "WARN: Few concise answer paragraphs. AI engines prefer 40-100 word direct answers."
    echo "Fix: Add a 1-2 sentence answer immediately after each question H2."
  fi
fi
echo ""

# 4. Page load speed (Core Web Vital proxy)
echo "--- Page Speed ---"
SPEED=$(curl -s -o /dev/null -w "%{http_code} %{time_total}s %{size_download}bytes" -L "$SITE" || true)
echo "HTTP $SPEED"
echo ""

# 5. Mobile-friendliness check
echo "--- Mobile Viewport ---"
VIEWPORT=$(curl -sL "$SITE" | grep -oP '<meta name="viewport"[^>]*>' || echo "MISSING")
echo "Viewport tag: $VIEWPORT"
echo ""

echo "=== SCORECARD ==="
SCORE=0
[ -n "$SD" ] && SCORE=$((SCORE + 25))
[ "$QUESTION_COUNT" -ge 3 ] && SCORE=$((SCORE + 25))
[ "$PCT" -ge 20 ] && SCORE=$((SCORE + 25))
[ -n "$VIEWPORT" ] && SCORE=$((SCORE + 25))
echo "AI SEO Readiness: ${SCORE}/100"
echo ""
echo "=== NEXT STEPS ==="
echo "1. Share this report with the client"
echo "2. Sell the full $1,500 Optimize package for the highlighted fixes"
echo "3. Build a retainer by proposing monthly monitoring ($2K/mo)"

Save as ai-seo-audit.sh and run: bash ai-seo-audit.sh https://client-site.com

The output is your deliverable — a live AI SEO readiness scorecard with prioritized fixes. Send it as a PDF to close the deal.

Deliverable Format

AI SEO Audit Report (Invoice-Ready Template)

Company: [Client Name]
URL: [https://client-site.com]
Date: [YYYY-MM-DD]
Auditor: [Your Name]

---

EXECUTIVE SUMMARY

AI search engines (Perplexity, ChatGPT Search, Google AI Overviews) now answer
queries directly in the SERP. [Client] currently scores [X]/100 on AI SEO readiness.
This report identifies [N] high-impact fixes that will increase citation rate and
brand visibility in AI-generated answers.

---

1. STRUCTURED DATA ANALYSIS
   Current: [JSON-LD present/missing/types found]
   Impact: [Missing schema = lost citations / Good schema = strong foundation]
   Fix: [Specific schema types to add with code example]

2. ANSWER FORMAT EVALUATION
   Current: [X question-format H2s found / concise paragraph ratio]
   Impact: [AI extraction difficulty score]
   Priority URLs to restructure:
   - [/page-1] — Rewrite H2 as question, add 40-word answer
   - [/page-2] — Add FAQPage schema
   - [/page-3] — Reformat list into direct-answer paragraph

3. TOPICAL AUTHORITY GAPS
   Missing subtopics that competing cited pages cover:
   - [Subtopic 1] — 0 articles vs 3 from competitor
   - [Subtopic 2] — No pillar page
   Recommendation: Create [N] new pages on above topics

4. TECHNICAL FOUNDATION
   - Page load: [seconds] — [pass/fail]
   - Mobile viewport: [present/missing]
   - Core Web Vitals proxy: [status]

5. PRIORITIZED FIX LIST
   Critical (do this week):
   1. [Fix] — [effort] — [expected impact]
   2. [Fix] — [effort] — [expected impact]
   Important (do this month):
   3. [Fix] — [effort] — [expected impact]
   4. [Fix] — [effort] — [expected impact]
   Growth (quarterly):
   5. [Fix] — [effort] — [expected impact]

---

PRICING

[X] Priority Fixes Implementation ........ $500
Full AI SEO Content Restructure (10 pages) ...... $1,500
Monthly Retainer (15 pages/mo + monitoring) .... $2,000/mo

---

APPENDIX: AI Engine Citation Scoring
- Perplexity: [cites/passes/blocked]
- ChatGPT Search: [visible/partial/missing]
- Google AI Overviews: [present/competing/absent]

Copy template, fill sections from audit output, brand with your logo, send as PDF.

Anti-Rationalization Table

Process

1. Research — Analyze target audience, competitors, and trending topics
2. Audit — Run ai-seo-audit.sh on target URLs, produce scorecard
3. Fix — Add JSON-LD schema, rewrite H2s as questions, add direct-answer paragraphs
4. Monitor — Track citation appearance in Perplexity, ChatGPT Search, Google AI Overviews
5. Report — Monthly citation report showing visibility gains and gaps closed

Verification

• All steps executed successfully
• Results validated against acceptance criteria
• Error handling tested with edge cases
• Documentation updated with findings