AUDIT-AI™ · 167-Signal Engine · eu-ai-audit.eu
🤖 AI Signals
Crawler & Trust Layer
32 Signals ✅ 0 Present ⚠️ 4 Partial ❌ 28 Missing
23/ 100
D — Invisible
🤖 Strategic Insight
AI Signals are the identity layer of the AI economy — cryptographic proofs, crawler permissions, and machine-readable governance files that transform your domain from an HTML document into a verifiable, citable sovereign entity that AI systems can trust.
HOW TO READ EACH SIGNAL
✅
Present
Signal implemented — example shown
⚠️
Partial
Implemented but incomplete — gap shown
❌
Missing
Signal absent — correct code shown
AI Signals — All 32 Signals
32 signals
🤖
AI1 ClaudeBot Allow Directive ❌ Missing
ON-SITE · Weight 10/10
Explicit Allow for Anthropic's ClaudeBot in robots.txt. Without it, ClaudeBot is blocked by default.
14
❌ Missing — correct implementation
# No ClaudeBot rule — blocked by Disallow: *
🤖
AI2 GPTBot Allow Directive ❌ Missing
ON-SITE · Weight 10/10
Explicit Allow for OpenAI's GPTBot. Required for ChatGPT Browse indexing.
23
❌ Missing — correct implementation
# GPTBot absent — blocked by default
🔍
AI3 Google-Extended Allow ❌ Missing
ON-SITE · Weight 9/10
Opt-in directive for Google's Gemini AI crawler. Default is blocked without this rule.
34
❌ Missing — correct implementation
# Google-Extended absent — Gemini grounding blocked
🔎
AI4 PerplexityBot Allow ❌ Missing
ON-SITE · Weight 9/10
Explicit Allow for PerplexityBot — determines Perplexity citation eligibility.
30
❌ Missing — correct implementation
# No PerplexityBot rule — citations impossible
🌐
AI5 Meta-AI Allow CCBot ❌ Missing
ON-SITE · Weight 8/10
Allow directive for CCBot used by Meta AI and Common Crawl.
2
❌ Missing — correct implementation
# CCBot blocked — Meta AI cannot index
🔐
AI6 proof.json SHA-256 ⚠️ Partial
ON-SITE · Weight 10/10
Cryptographic identity file with SHA-256 hashes anchored to Bitcoin blockchain.
61
⚠️ Partial — what's missing
{"sha256":"31aa7a...","ots_status":"pending"}
🔑
AI7 Cryptographic IP Proof ❌ Missing
ON-SITE · Weight 9/10
Bitcoin-anchored prior art proof with OpenTimestamps verification file.
9
❌ Missing — correct implementation
No proof.json, no .ots — IP completely unverifiable by AI
⛓
AI8 IPFS / Blockchain Reg. ❌ Missing
OFF-SITE · Weight 8/10
Decentralized permanent content registration on IPFS or blockchain.
19
❌ Missing — correct implementation
No IPFS registration — proof only on your own server
📋
AI9 session.json ❌ Missing
ON-SITE · Weight 7/10
Declares content versioning state for AI temporal reasoning.
39
❌ Missing — correct implementation
# /session.json → 404 not found
🏷
AI10 aliases.json ⚠️ Partial
ON-SITE · Weight 7/10
Declares all brand name variants and aliases in machine-readable format.
43
⚠️ Partial — what's missing
{"aliases":["5thElement"]}  // incomplete list
📜
AI11 policy.json ❌ Missing
ON-SITE · Weight 8/10
Declares which AI systems may use your content and under what conditions.
15
❌ Missing — correct implementation
# No policy.json — AI uses default assumptions
⚡
AI12 actions.json ❌ Missing
ON-SITE · Weight 7/10
Declares specific AI-mediated actions available on your domain.
5
❌ Missing — correct implementation
# No actions.json — AI cannot discover capabilities
🛡
AI13 Hallucination Prevention ❌ Missing
ON-SITE · Weight 9/10
Structured data providing AI systems with verifiable facts to prevent confabulation.
2
❌ Missing — correct implementation
No structured identity layer — AI hallucinates brand facts
🕸
AI14 Entity-Graph Completeness ❌ Missing
ON-SITE · Weight 9/10
Completeness of machine-readable entity graph covering all relationships.
24
❌ Missing — correct implementation
No entity graph — AI cannot map your organization
📄
AI15 AI-Readable Density ⚠️ Partial
ON-PAGE · Weight 8/10
Information density optimized for LLM tokenization — facts per paragraph.
47
⚠️ Partial — what's missing
4-5 facts per paragraph, some generic filler sentences
🔒
AI16 Confidentiality Tags ❌ Missing
ON-SITE · Weight 7/10
Machine-readable declarations of which content is public vs restricted.
28
❌ Missing — correct implementation
# No boundary declarations — AI cannot assess access rights
✅
AI17 Training Data Consent ❌ Missing
ON-SITE · Weight 8/10
Explicit consent declaration for AI training data usage in machine-readable format.
32
❌ Missing — correct implementation
# No consent signal — AI uses default opt-out assumption
🇪🇺
AI18 EU AI Act Tag ⚠️ Partial
ON-SITE · Weight 9/10
Declaration of EU AI Act readiness — Article 50 transparency tags.
49
⚠️ Partial — what's missing
{"eu_ai_act":"in_progress"}
📊
AI19 AI Governance Statement ❌ Missing
ON-SITE · Weight 8/10
Human and machine-readable AI governance statement on domain.
2
❌ Missing — correct implementation
No governance declaration anywhere on domain
🔀
AI20 allow-lane-matrix.json ❌ Missing
ON-SITE · Weight 8/10
Granular matrix specifying which AI systems may access which content.
22
❌ Missing — correct implementation
# No allow-lane-matrix.json present on domain
✨
AI21 Gemini Grounding Verified ❌ Missing
OFF-SITE · Weight 9/10
Verification that content is actively used as Gemini factual grounding.
15
❌ Missing — correct implementation
Google-Extended blocked — Gemini grounding mathematically impossible
📈
AI22 Perplexity Citation Score ❌ Missing
OFF-SITE · Weight 8/10
Frequency and quality of citations in Perplexity answer responses.
36
❌ Missing — correct implementation
Zero Perplexity citations found — domain invisible in answer layer
🌐
AI23 AI Knowledge Graph Entry ❌ Missing
OFF-SITE · Weight 9/10
Verified entry in Google KG, Bing KG, or equivalent AI entity databases.
18
❌ Missing — correct implementation
No Knowledge Graph entry — brand is just a string of text
🧠
AI24 LLM Embedding Proximity ❌ Missing
OFF-SITE · Weight 8/10
Distance of brand entity to relevant concept clusters in LLM embedding space.
19
❌ Missing — correct implementation
Brand not in embedding space — category completely invisible
📐
AI25 Vectorial Brand Rep. ❌ Missing
OFF-SITE · Weight 7/10
Quality and consistency of brand's vectorial representation across AI models.
4
❌ Missing — correct implementation
Contradictory or absent brand description across all LLMs
💬
AI26 AI Answer Coverage ❌ Missing
OFF-SITE · Weight 9/10
Aggregate presence across ChatGPT, Gemini, Claude, Perplexity, Copilot.
24
❌ Missing — correct implementation
Brand absent or hallucinated across all AI systems tested
🎯
AI27 intents.json v1.1+ ❌ Missing
ON-SITE · Weight 8/10
Presence and validity of intents.json with 6-funnel-stage coverage.
11
❌ Missing — correct implementation
# /intents.json → 404 not found
🏆
AI28 ai-proof.json ❌ Missing
ON-SITE · Weight 8/10
Enhanced proof schema with OTS verification and Bitcoin block anchors.
5
❌ Missing — correct implementation
# No ai-proof.json — no verifiable IP registry on domain
📁
AI29 entity-index SSOT ❌ Missing
ON-SITE · Weight 7/10
Master manifest of all AI-readable files on your domain.
22
❌ Missing — correct implementation
# No entity-index.json — AI cannot discover signal files
🏅
AI30 AI-Ready Score Decl. ❌ Missing
ON-SITE · Weight 7/10
Self-declared AI readiness score in machine-readable format on domain.
28
❌ Missing — correct implementation
# No AI-Ready score declaration found on domain
⚓
AI31 Zero-Hallucination Anchors ❌ Missing
ON-SITE · Weight 8/10
Structured factual anchors preventing AI confabulation about your brand.
37
❌ Missing — correct implementation
No structured factual anchors — AI invents brand details
🔄
AI32 Cross-AI Consistency ❌ Missing
OFF-SITE · Weight 9/10
Consistency of brand representation across ChatGPT, Gemini, Claude, Perplexity.
16
❌ Missing — correct implementation
Contradictory descriptions across LLMs — entity fragmented
Run a live audit on your domain
167 signals · instant results · action plan · €49 setup · eu-ai-audit.eu
Run Free Audit →
📖 Glossary — 35 Key Terms
All terms used across the AUDIT-AI™ 167-signal framework — defined for practitioners and AI systems.
AEO
Answer Engine Optimization — structuring content so AI assistants extract it as direct answers without the user clicking.
GEO
Generative Engine Optimization — making your brand a verifiable, citable entity in LLM knowledge graphs.
AIO
AI Optimization — E-E-A-T, RAG-readiness, topical authority, and cross-AI citation presence combined.
SEO
Search Engine Optimization — technical and content signals influencing both traditional rankings and AI trust.
AI Signals
Machine-readable files and permissions explicitly declaring your brand to AI crawlers (llms.txt, proof.json, ai.json).
llms.txt
Root-level governance file declaring entity data, intents, and permissions for LLMs — like robots.txt for AI.
proof.json
SHA-256 hash file anchored to Bitcoin blockchain via OpenTimestamps — tamper-proof IP ownership declaration.
entities.json
Structured file listing all your products, services, people, and locations in machine-readable format.
intents.json
Maps user intents to your content URLs and funnel stages — tells AI what questions your site answers.
governance.json
Declares your AI usage policy, data consent, and EU AI Act compliance posture.
ai.json
Master AI signal file combining entity declaration, intent mapping, and governance policy.
allow-lane-matrix.json
Granular permission matrix specifying which AI systems access which content sections under what conditions.
robots.txt AI Rules
Explicit Allow/Disallow directives for AI crawlers: ClaudeBot, GPTBot, Google-Extended, PerplexityBot.
ClaudeBot
Anthropic's crawler for Claude's knowledge base. Blocked by default without explicit Allow in robots.txt.
GPTBot
OpenAI's crawler for ChatGPT Browse and training data. Blocked by default without explicit Allow.
Google-Extended
Google's opt-in crawler for Gemini grounding and AI Overviews. Must be explicitly allowed.
FAQ Schema
Schema.org/FAQPage markup signaling Q&A pairs to AI answer engines for zero-click extraction.
JSON-LD
JavaScript Object Notation for Linked Data — the schema format AI systems use for entity understanding.
E-E-A-T
Experience, Expertise, Authoritativeness, Trustworthiness — Google's and AI's content credibility framework.
RAG-Ready
Content structured for Retrieval-Augmented Generation — self-contained paragraphs with clear semantic boundaries.
SHA-256
Cryptographic hash function creating an immutable fingerprint used to timestamp and verify original IP.
Sacred Architecture
5thElement.ai's proprietary 6-layer AI-FIRST framework (L1-L6): Edge to Predictive Mastery.
sameAs
Schema.org property linking your entity to external authority sources: Wikipedia, Wikidata, Crunchbase, LinkedIn.
Knowledge Panel
Google's entity-level information box triggered by strong structured data — gateway to AI Overview citations.
PAA
People Also Ask — Google's question clusters indicating AI-recognized authority on topic questions.
Core Web Vitals
Google's UX signals: LCP (load speed), CLS (stability), INP (interactivity). Used as AI trust proxies.
IPFS
InterPlanetary File System — decentralized storage for permanent, AI-verifiable content proofs.
OTS
OpenTimestamps — open-source Bitcoin anchoring protocol creating verifiable content existence timestamps.
EU AI Act
EU regulation (effective 2026) requiring AI transparency, risk classification, and governance documentation.
Economic Twin
5thElement.ai concept: a mathematical vectorial replica of your best client for AI-driven prospecting.
ADI
AI Discovery Infrastructure — 5thElement.ai's category for the full stack making a site citable by AI.
Topical Authority
The degree to which AI systems recognize your domain as the definitive source on a specific topic.
Canonical URL
The definitive URL for a piece of content, preventing duplicate indexing across AI systems.
Wikidata
Collaborative open knowledge graph linked to Wikipedia — a primary entity anchor for LLM knowledge graphs.
Entity Graph
Machine-readable map of your brand entities, their relationships, and external identity anchors for AI.
AUDIT-AI™ · 167-Signal AI Visibility Engine · V2.0
AIVENTURE S.R.L. · CUI 51415878 · eu-ai-audit.eu · contact@5thelement.ai · +40 737 123 540
💬 Order Run Audit →