Legible.
The full system that makes a brand legible to AI

Legibility, engineered.

Twelve modules, one job — making your brand findable, citable, and buyable by AI agents, and proving it.

Pricing →
How an AI agent reads your site

Most sites are invisible to agents.

Without Legible
<div id="root"></div>
<script src="/app.js"></script>
Entity
unknown
Extractable claims
none
Cited in AI answers
no
With Legible
{ "@type": "Organization",
  "name": "Acme Corp",
  "sameAs": ["wikidata.org/Q…"],
  "areaServed": "EU" }
Entity
Acme Corp · grounded ✓
Extractable claims
14 · supported
Cited in AI answers
yes

Illustrative example

Fix Pack

Generated fixes, not advice.

Copy-paste JSON-LD, llms.txt, robots 2.0 and ai-facts — pre-filled for your site.

  • JSON-LD entity & product schema
  • llms.txt + robots 2.0
  • ai-facts.json truth file
fix-pack.jsonld
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Acme Corp",
  "url": "https://acme.com",
  "sameAs": [
    "https://www.wikidata.org/wiki/Q…",
    "https://www.linkedin.com/company/acme"
  ],
  "areaServed": "EU",
  "foundingDate": "2019"
}

Illustrative example

Share-of-Voice

Are you in the answer?

Are you named in AI answers? Measured across Claude, ChatGPT, Perplexity & Gemini.

“best agent-readiness platform?”
1Competitor A
92%
2Competitor B
64%
3Competitor C
48%
4Your brand
18%
ChatGPT · Claude · Gemini · PerplexityIllustrative example
Entity grounding

One identity, everywhere.

Wikidata + sameAs identity graph so models agree on who you are.

Your brand
Wikidata
Wikipedia
LinkedIn
Crunchbase
sameAs · verified P856 match

Illustrative example

What the index measures8 dimensions · 100 pts
  • Structured data
    Valid JSON-LD across Organization, Product, FAQ — the one lever with proven citation lift.
  • Entity consensus
    One canonical description and a sameAs identity graph, identical everywhere.
  • Citability
    Atomic, extractable claims a model can quote in a single sentence.
  • Commerce readiness
    Machine-readable price, stock and policy — can an agent actually buy from you?
  • Machine access
    Fast, render-free data; AI crawlers allowed; a sitemap to follow.
  • Multilingual
    hreflang and per-locale parity — the sources AI cites differ by language.
  • Verification
    Claims backed by evidence and a clear publisher — the antidote to AI-washing.
  • Freshness
    Recency signals; citation churn punishes stale data.
The platform
01
Readiness Index
An 8-dimension score with band, multi-page crawl and platform detection.
02
Fix Pack
Copy-paste JSON-LD, llms.txt, robots 2.0 and ai-facts — pre-filled for your site.
03
Entity grounding
Wikidata + sameAs identity graph so models agree on who you are.
04
Hosted truth layer
A machine-readable doc agents read at a stable URL, kept current.
05
Share-of-Voice
Are you named in AI answers? Measured across Claude, ChatGPT, Perplexity & Gemini.
06
Per-locale visibility
DE / FR / IT / EN parity — visibility differs sharply by language.
07
Deep Audit
An agent's-eye qualitative read of legibility, claims and transactability.
08
Verification
Claims graded supported / vague / unsupported — your anti-AI-washing score.
09
Commerce readiness
Can an agent complete a purchase? A callable audit says yes / partial / no.
10
Agent-commerce manifest
Your machine-readable catalogue, pricing, availability and policy surface.
11
Protocols
ACP / UCP / MCP readiness so agents can transact, not just read.
12
Agent governance
Identify & verify AI crawlers (FCrDNS) into trust tiers; log every agent read.
How it worksscore · fix · serve · measure
01

Score

Paste a URL. We read your public site and grade 8 machine-facing dimensions into one Readiness Index.

02

Fix

Get a prioritized Fix Pack — copy-paste JSON-LD, llms.txt, robots and ai-facts, pre-filled for your site.

03

Serve

Publish a hosted truth layer and agent-commerce manifest that agents read directly, kept fresh.

04

Measure

Track Share-of-Voice across models and languages, verify claims, and watch verified agent traffic.

See where your brand stands with AI.

Get your Readiness Index in seconds — then the Fix Pack to close the gaps.

→Pricing