--- title: "KYE™ Site Authority™ — scan your site's authority for AI agents · KYE Protocol™" description: "KYE™ Site Authority™: a free, non-fakeable site scan that scores how a site discloses its authority and evidence posture to AI agents — across six dimensions (content authority, publishing evidence, agent readiness, change governance, action authority, trust registry). Scored live in your browser from observable signals only. SiteTrust certifies AI transparency. Cloudflare checks agent-readiness. KYE™ verifies site authority for AI agents." url: https://kyeprotocol.com/site-authority/ lang: en source: "KYE Protocol" --- > KYE™ Site Authority™: a free, non-fakeable site scan that scores how a site discloses its authority and evidence posture to AI agents — across six dimensions (content authority, publishing evidence, agent readiness, change governance, action authority, trust registry). Scored live in your browser from observable signals only. SiteTrust certifies AI transparency. Cloudflare checks agent-readiness. KYE™ verifies site authority for AI agents. # Scan your site's authority for AI agents **KYE™ Site Authority™** is the web/publisher channel of [Governance for the AI Web™](https://kyeprotocol.com/governance-for-the-ai-web/): a free, **non-fakeable** scan, a public registry, and an embeddable badge. As AI agents read and act on the web, the question shifts from _"is my site trustworthy?"_ to _"does my site disclose its authority so agents can verify it?"_ The scan scores that across six dimensions from **observable signals only**. You raise your score by _publishing the signal_, not by asserting you are good. It runs in your browser — nothing is sent anywhere. Scan a site now [See the product family](https://kyeprotocol.com/governance-for-the-ai-web/) ## Where KYE™ sits — SiteTrust, Cloudflare, KYE™ Three different jobs on the agent-readable web. They complement, they don't compete: ### SiteTrust Certifies AI transparency — how a site discloses its use of AI to people. ### Cloudflare Checks agent-readiness — whether infrastructure serves AI crawlers and agents. ### KYE™ Site Authority™ Verifies — does the site disclose, with evidence, who is speaking with authority and how that authority changes over time. site authority for AI agents **The red line — never a truth score.** KYE™ Site Authority™ asserts only verifiable facts: _evidenced · governed · disclosed · authority-exists_. It never outputs a truth score, a trust score, or a "true/false" verdict about a site or its operator. Privacy-by-design: the scan evaluates only public, site-published signals — never your visitors' browsing data. ## Run the scan Tick the signals your site publishes — or load KYE™'s own published set to see a fully-disclosed **authoritative** posture over signals you can fetch and confirm yourself. Each row names the observable signal and what publishing it attests. ## Contact Your request goes to the KYE Protocol™ team, and a receipt with a reference number is emailed to you. Name Email Organisation Topic Message I agree to the [privacy policy](https://kyeprotocol.com/legal/privacy/). Send ## Your Site Authority Score Site Authority (0–100) **This is an OBSERVED score** — a client-side heuristic over the signals you ticked, with no registry backing. A **VERIFIED** score is a registry-backed **KYE™ Site Authority Registry™** record: a canonical site UUID, a [Evidence Pack™](https://kyeprotocol.com/evidence-pack/), and a WORM change-history ledger ("archive.org for authority, not snapshots"). Observed is a snapshot; verified is authority. ### The six dimensions ### Your gaps — the signals to publish next Each gap names a concrete, observable signal that was absent and what publishing it would attest. Closing a gap is a verifiable act, not a claim. ### Claim your KYE™ Site Authority™ record Turn this observed snapshot into a public, lookup-able, registry-backed record with an embeddable badge. Claiming goes through the same **Universal Engagement Rail** every KYE™ applicant uses — you choose exactly what to share. [Claim your KYE™ Site Authority™ record](https://kyeprotocol.com/poc/) The scan runs entirely client-side (scorer ); nothing leaves your browser until you choose to claim a record. ## The registry + the embeddable badge A verified record is published to the **KYE™ Site Authority Registry™** — a [Directory](https://kyeprotocol.com/directory/) projection anyone can look up by the site's canonical UUID. The site UUID _is_ the canonical `kye::::` ID (reused, never forked). Each record carries a public [record page](https://kyeprotocol.com/site-authority/record/) and the embeddable badge: ``` ``` The badge renders the **live** registry posture and links to the public record — it never out-lives the record (Authority Finality™). A revoked or expired record flips the badge off. Its machine-readable metadata is honest by construction: `{kye_site_id, entity, authority_status, agent_readiness_score, publishing_authority_score, last_verified, evidence_registry}` — never a truth score. ## What's verified today, and what's next P1 ships the real scanner, the seeded registry, and the badge — and it tells you honestly what runs now versus what is deferred. We never fake a capability. - **Today (P1)** The scan runs the real, deterministic scoring engine over the signals you submit or over KYE™'s own published set — six dimensions, specific gaps, honest observed-vs-verified labelling, a seeded registry, the embeddable badge, and a public record page. - **Phase 2** A browser extension and a live, SSRF-safe scan of an arbitrary remote domain (fetching that site's llms.txt / robots.txt / metadata server-side) — declared honestly as deferred external runtime rather than faked. Canonical KYE™ surfaces referenced on this page: [KYE Protocol™](https://kyeprotocol.com/) · [Universal Engagement Rail](https://kyeprotocol.com/glossary/).