# 2026 AEO Checklist, llms.txt Validator & Deterministic AEO Grader Specification

> **Pillar:** Technical SEO · **Evidence Tier:** 🔵 Open Protocol / RFC Spec · **Updated:** 2026-11-06
> **Canonical Source:** AEORank Open Knowledge Hub (https://aeorank.app/learn/aeo-checklist-llms-txt-validator-grader)
> **Vietnamese Edition:** https://aeorank.app/tri-thuc/bang-kiem-aeo-checklist-va-llms-txt-validator

## Answer-First Summary
The 2026 AEO Checklist, anti-soft-404 llms.txt validator, and deterministic AEO grader (AEO checker) form a 62-rule technical specification across 9 weighted pillars designed to verify website readiness for Google AI Overviews, ChatGPT Search, Perplexity, and Claude. This engineering guide details the exact pillar weights, the 4-layer inspection logic for validating /llms.txt without SPA fallback errors, and the 0-to-100 scoring formula deployed on the AEORank Diagnostic Station.

## Key Takeaways & Epistemic Verification
- 🟢 [9-Pillar Technical AEO Checklist]: Unifies crawlability (robots.txt, Sitemap CFA-02), Unicode NFC meta bounds (AEO-002), Protected-Field year alignment (PFA-01), and self-contained 160-to-1,100-character leaf sections.
- 🔵 [Anti-Soft-404 llms.txt Validator & RFC 9727]: Verifies direct HTTP 200 OK status, text/plain or text/markdown MIME types, valid H1/H2 Markdown structure, and rejects single-page application (SPA) HTML soft-404 fallbacks.
- 🟡 [Deterministic AEO Grader & Checker Formula]: Combines 62 deterministic rules with 18 typed probabilistic AEO-Sem signals (enforcing a decisive p >= 0.80 threshold) to compute reproducible 0-to-100 readiness scores.

## 1. The 9-pillar deterministic AEO checklist for technical and content teams

A production-grade 2026 AEO checklist must audit the complete stack from HTTP response headers and XML Sitemap timestamps down to semantic DOM passage boundaries. Even if a page features well-written copy, it will fail AI retrieval gates if an overly broad robots.txt rule blocks OAI-SearchBot, if uniform Sitemap lastmod dates trigger rule CFA-02, or if title tags exceed 65 Unicode NFC characters and truncate on the SERP.

The matrix below outlines how the AEORank AEO checker and AEO grader weight all 9 evaluation pillars (normalizing 115 raw weight points to a 0-to-100 scale) so engineering and editorial teams can prioritize fixes systematically.

| AEO Checklist Pillar | Normalized Weight | Core AEO Checker Rules | 100/100 AEO Grader Passing Criteria |
| --- | --- | --- | --- |
| 1. Crawlability & Sitemap Integrity | 17.4% (20/115) | HTTP 200, Canonical alignment, authentic <lastmod> distribution (CFA-02) | Zero orphan pages; < 80% uniform Sitemap lastmod dates |
| 2. Content & Semantic Quality (AEO-Sem) | 17.4% (20/115) | 500-char Answer-First lead, self-contained 160–1,100 char leaf sections | sections.weak == 0; answer_first & citable >= 0.80 |
| 3. AI Search & Agentic Protocols | 17.4% (20/115) | 10 AI Crawlers allowed (RFC 9309), /llms.txt validator, RFC 9727 catalog | 6/6 Agentic Web signals & verified live AI citations |
| 4. On-Page, Schema & Security Stack | 47.8% (55/115) | Title 15–60 NFC chars, Meta 100–160 chars, PFA-01, Schema sameAs, HSTS | Zero PFA-01 year conflicts; >= 4 verified sameAs nodes |

## 2. How an anti-soft-404 llms.txt validator and RFC 9727 checker work

The most frequent implementation bug on modern React, Vue, Next.js, and SPA websites is the Soft-404 fallback trap: when an AI crawler requests `/llms.txt`, the web server returns `HTTP 200 OK` but serves the entire HTML homepage bundle instead of a Markdown file. Superficial checkers that only inspect the HTTP status code falsely mark the file as valid, whereas GPTBot, ClaudeBot, and PerplexityBot encounter raw HTML tags and abort context ingestion immediately.

AEORank's built-in llms.txt validator performs a strict 4-layer check: verifying `HTTP 200 OK` without redirect loops, confirming a `text/plain` or `text/markdown` Content-Type header, rejecting payloads that start with `<!DOCTYPE html>` or `<html`, and validating the presence of a top-level `# H1` title, a `>` summary blockquote, and canonical Markdown links paired with `/.well-known/api-catalog` (RFC 9727).

## 3. AEO grader vs. surface checkers: Weighted scoring and readiness thresholds

Unlike surface-level SEO checkers that merely count heading tags or keyword frequency, a true AEO grader blends deterministic protocol verification with typed semantic probability evaluation. Within the AEORank scoring engine, the Content and AI Search pillars combine 30% deterministic rule compliance with 70% probabilistic evaluation from the AEORank Semantic Engine (AEO-Sem, published under CERN Zenodo DOI 10.5281/zenodo.23582947), counting a semantic signal as passed only when its confidence reaches `p >= 0.80`.

Every deduction in the AEO grader maps directly to an actionable remediation snippet inside the automated 6-File Patch Pack—including a ready-to-deploy `/llms.txt` index, an additive RFC 9309 `robots.txt` block, and XSS-safe JSON-LD Schema. Enter your URL in the inline diagnostic scanner below or on the AEORank homepage to run the live AEO checker, AEO grader, and llms.txt validator in 15 seconds.

## Frequently Asked Questions (AEO & People Also Ask)

### 1. What are the highest-priority items in a 2026 technical AEO checklist?
The top P1 priorities in the 2026 AEO checklist are allowing all 10 AI search crawlers via additive RFC 9309 robots.txt rules, publishing a verified /llms.txt context index alongside /.well-known/api-catalog (RFC 9727), placing a 40-to-60-word Answer-First definition within 500 characters below H1, and eliminating Protected-Field year conflicts (PFA-01).

### 2. How does an llms.txt validator detect SPA soft-404 fallback errors?
Instead of relying solely on an HTTP 200 status code, AEORank's llms.txt validator inspects the Content-Type header for text/plain or text/markdown, rejects response bodies containing <!DOCTYPE html> or <html tags from SPA routers, and verifies valid Markdown # H1 headings, blockquotes, and canonical links.

### 3. How does an AEO grader calculate a deterministic 0-to-100 readiness score?
The AEORank AEO grader normalizes 9 weighted pillars totaling 115 raw weight points onto a 0-to-100 scale (Overall = round(sum(pillar_score * weight_raw) / 115)), blending 30% deterministic rule checks with 70% AEO-Sem typed semantic probability judgments (p >= 0.80) on the Content and AI Search pillars.

---
## Editorial Board & Primary Sources (E-E-A-T)
- **Authored & Reviewed by:** Principal Engineer Le Khoa & AEORank (developers of the AEORank v2.1 audit engine combining 62 deterministic rules with 18 `AEO-Sem` semantic probability judgments).
- **Primary References:** Google Search Central Documentation (`developers.google.com/search/docs`), IETF RFC 9727 (`rfc-editor.org/rfc/rfc9727`), KDD 2024 GEO Benchmark (`arxiv.org/abs/2311.09735`), and `llmstxt.org`. Contact: `support@aeorank.app`.

## Run a Live Diagnostic on Your Domain
- Free 15-second 62-rule SEO, AEO & GEO diagnostic scanner: [AEORank Global Diagnostic Station](https://aeorank.app/)
- Machine-readable AI context directory: [llms.txt](https://aeorank.app/llms.txt) · Full RAG Corpus: [llms-full.txt](https://aeorank.app/llms-full.txt)