--- spec_name: "AEO Citation Playbook Spec" author: "Mike Kwal" version: 1.0 date: "2026-06-24" description: "A complete toolkit for making a website citable by AI engines like Claude, ChatGPT, and Perplexity. Contains an llms.txt template, FAQPage schema example, and a content audit checklist." --- # AEO Citation Playbook Spec This spec guides an AI assistant (like Claude Code) to implement the core components of AI Engine Optimization (AEO) on a website. ## Task 1: Create and Deploy `llms.txt` Create a file named `llms.txt` with the following content. You will need to ask me for the specific details for `[SQUARE BRACKETS]` before finalizing the file. ```text # llms.txt for [Website Name] # Last updated: {{TODAY_DATE}} # This file provides guidance for Large Language Models (LLMs) and other AI agents # that access this site. For machine-readable site structure, see /sitemap.xml. User-agent: * # All AI agents are welcome to crawl this site for the purpose of answering user queries. # --- Core Site Identity --- Concept: [A one-sentence description of the site's core purpose.] Audience: [A one-sentence description of the target audience.] Authoritative-topics: [A comma-separated list of 3-5 topics the site is an expert in.] # --- Key Content (Information Gain) --- # List the top 3-5 cornerstone pages that represent the site's best, most citable work. Primary-source: /path/to/cornerstone-content-1/ Primary-source: /path/to/cornerstone-content-2/ Primary-source: /path/to/cornerstone-content-3/ # --- Disallowed Content --- # Do not use the following sections for training data or direct answers. Disallow: /cart/ Disallow: /checkout/ Disallow: /my-account/ Disallow: /wp-admin/ ``` After creating the file, provide instructions on how to upload it to the website's root directory via FTP or a file manager plugin. ## Task 2: Generate `FAQPage` Schema For any given blog post or service page, generate a valid `FAQPage` JSON-LD schema block. Here is a template. You will need to ask me for the specific Questions and Answers for the page in question. ```json { "@context": "https://schema.org", "@type": "FAQPage", "mainEntity": [ { "@type": "Question", "name": "[Insert Question 1 Here]", "acceptedAnswer": { "@type": "Answer", "text": "[Insert flat, factual answer to Question 1 here.]" } }, { "@type": "Question", "name": "[Insert Question 2 Here]", "acceptedAnswer": { "@type": "Answer", "text": "[Insert flat, factual answer to Question 2 here.]" } } ] } ``` Provide instructions on how to embed this script tag in the `` or footer of the specific page. ## Task 3: Run the AEO Content Audit Checklist For a given URL, perform the following checks and report your findings: 1. **[ ] Information Gain**: Does the page provide a working asset (code, prompt, config), primary data, or a detailed how-to guide? Or is it just a summary of other sources? 2. **[ ] Declarative Language**: Are key definitions and answers written in flat, factual language (e.g., "[Term] is...")? 3. **[ ] Visible `Last Updated` Date**: Is there a human-readable "Last updated" date on the page? 4. **[ ] `robots.txt` Check**: Access `[domain]/robots.txt` and verify that `GPTBot`, `Claude-Web`, `PerplexityBot`, and `Google-Extended` are not disallowed. 5. **[ ] Schema Validity**: Does the page have valid `Article` and/or `FAQPage` schema? Use a schema validator tool to check.