← All skills
SKILL

Search + AI Readiness

GOOGLE + AI

PEOPLE-FIRST PUBLISHING WITH CURRENT GOOGLE GUIDANCE, SOURCE FIDELITY, CRAWLABILITY, AND AN INTERNAL AI CITATION READINESS HEURISTIC. NO GUARANTEED RANKINGS OR CITATIONS.

By Somanath Diksangi * Last updated: July 29, 2026 * Source: Google AI Optimization Guide 2026

COMMAND
$ /claudeblog geo <file>

REQUIRES CLAUDE SUITE INSTALLED IN CLAUDE CODE

HOW IT WORKS

What does Search + AI Readiness check?

After writing, run /claudeblog geo on your draft. It scores AI citation readiness across 5 evidence-backed dimensions: Evidence-Backed Citability (27 pts), Purpose Fit & Reader Utility (20 pts), Entity Clarity (20 pts), Content Structure for Extraction (20 pts), and AI Crawler Accessibility (13 pts). Google's official position (June 2026): GEO is rebranded SEO — no special AI files, llms.txt, or content chunking is needed for Google's AI features. The same core ranking signals power AI Overviews, AI Mode, and classic Search. Optimize passage clarity, source fidelity, and entity structure.

Each stage feeds into the next, creating a complete content production pipeline. The Reviewer scores the final output and ensures it meets the quality gate before delivery.

STAGES
01

Evidence-Backed Citability

Self-contained passages (134-167 words) with specific claim + supporting evidence + source attribution. ~44% of AI citations come from first 30% of a page.

02

Purpose Fit & Utility

Clear page purpose identified in introduction, useful section openings without throat-clearing, format matched to intent (declarative headings, questions, FAQs, tables).

03

Entity Clarity

One unambiguous primary topic per page, consistent naming throughout, clear topic statement in intro, title accurately reflects content.

04

Content Structure

Standalone summary box, comparison tables with semantic headers (<thead>), ordered lists for processes, definition formatting for key terms, evidence-backed explanations for important claims.

05

AI Crawler Accessibility

Content present in rendered DOM, Google indexability is clean, robots.txt policies match AI visibility goals, schema reaches rendered DOM, reasonable page size.

SCORING

How are issues categorized?

Each category produces its own score out of 100, with a pass/warn/fail status. The overall score is computed from all categories combined. Issues are grouped by severity for clear prioritization.

SeverityTimeline
Poor (0-49)Major restructure needed
Needs Work (50-69)Significant gaps
Good (70-89)Minor improvements needed
Excellent (90-100)Highly citable by AI systems
USAGE

How do you run Search + AI Readiness?

01

Write a draft

/claudeblog write "your topic"
02

Run GEO check

/claudeblog geo ./path/to/draft.md
03

Optimize passages

Follow the citability recommendations and evidence improvements
FAQ
Is this guaranteed to get AI citations?▼
No. The GEO audit provides a readiness heuristic (0-100), not a calibrated citation probability. The score is an editorial tool, not a prediction. Google's own docs (June 2026) confirm no tool can guarantee citations.
Does it cover both Google and non-Google AI?▼
Yes. The audit checks Google eligibility (normal crawlability/indexability — no special markup needed) AND non-Google AI crawlers (GPTBot, ClaudeBot, PerplexityBot, etc.). Platform-specific recommendations are provided for ChatGPT, Perplexity, AI Overviews, AI Mode, Claude, Gemini, Copilot, and You.com.
What's the optimal passage length for AI citations?▼
134-167 words for self-contained answer blocks. However, Google prescribes no passage-length or chunking requirement. Citations correlate with clarity and evidence, not word count bands.
Does llms.txt help with Google AI citations?▼
No. Google explicitly states (AI Optimization Guide, June 2026) that llms.txt 'won't harm (nor help) your visibility or rankings in Google Search, as Google Search ignores them.' John Mueller separately called the llms.txt discovery use case 'a dead end.' It may help non-Google AI services.

Run Search + AI Readiness in one command

$ /claudeblog geo <file>