AXIOM — Agent eXecution, Information & Orchestration Markup
Your site has two interfaces.You only built one of them.
Frontier AI agents can afford to render your page in a headless browser. They're a rounding error. The actual agentic web is thousands of simple agents making raw HTTP requests and parsing whatever comes back. Your HTML is their only interface. If it's empty, broken, or illegible, they move on.
Two Documents. One Standard.
The Scoring Framework tells you where the gaps are. The Build Spec tells you how to close them.
Scoring Framework
Measurement & Diagnostics
The measurement system. Six dimensions, quantitative scoring, letter grades. Produces audit reports with severity-rated findings and prioritized remediation.
- Purpose
- Diagnostic scoring
- Audience
- Site owners, auditors, consultants
- Question
- “How agent-ready is this site?”
Build Spec
Implementation Standard
The implementation standard. Traffic governance, agent manifest, markup vocabulary, readiness levels. Defines how to build sites that AI agents can actually use.
- Purpose
- Implementation standard
- Audience
- Developers, framework authors, CMS builders
- Question
- “How do I build an agent-ready site?”
What AXIOM Is Not
Every standard below is valuable. None of them answer the question AXIOM answers: when an agent fetches your page, can it find the content, understand the structure, locate the actions, and extract the data?
| Standard | What It Solves | What It Doesn't |
|---|---|---|
| MCP | How an LLM connects to external tools | Whether your page is readable when fetched |
| A2A | How agents communicate with each other | Whether your page is navigable by an agent |
| AGENTS.md | Developer-facing instructions in markdown | Machine-readable governance and traffic policy |
| GEO | Whether AI models cite your brand in answers | Whether agents can operate and transact on your site |
| WCAG | Whether humans with assistive tech can use your site | Whether machines with no human in the loop can use your site |
| AXIOM | Page-level machine readability + structured agent governance | — |
Six Dimensions of Agent Readiness
AXIOM scores your site across six dimensions, weighted by impact. A site that fails on Content Survivability can't score well on anything downstream.
Content Survivability
25%Does content exist without JavaScript?
Structural Legibility
20%Can agents identify the page layout from the DOM?
Interactive Manifest Clarity
20%Can agents find and invoke your CTAs?
Data Extractability
15%Is business data in semantic, parseable HTML?
Navigation Traversability
10%Can agents explore via static links?
Agent Response Fitness
10%Does the text stream tell a coherent story?
Why Agent Readiness Matters
AI agents are already browsing, comparing, and purchasing on behalf of customers. Sites that can't serve them are losing revenue to competitors that can.
The Silent Failure
When an agent visits a modern SPA, it doesn't open a browser. It reads the raw HTML. If that HTML is an empty <div id="app"></div> shell, the agent sees nothing — no products, no prices, no purchase buttons. It moves to a competitor whose HTML contains actual content. No error log, no bounce event. Just a sale that went somewhere else.
The Inverted Funnel
Human conversion is driven by visual design and emotional appeal. Agent conversion is driven by structured data, action discoverability, and information reliability. A site optimized exclusively for human visitors can be completely non-functional for agent visitors.
The robots.txt Crisis
AI companies use the same user-agent for training crawlers and customer-facing agents. Block the crawler and you block your customers' agents too. robots.txt can't distinguish them. AXIOM's axiom.json can — one JSON file, deployed in five minutes.
Accessibility Is Not Enough
WCAG compliance helps humans with assistive technology. Agent readiness helps machines with no human in the loop. A fully WCAG-compliant React app can score zero on agent readiness if it returns an empty page without JS.
Read the Documents
The business case explains why this matters. The specs explain how it works.
Business Case
The Agentic Web
Why your website is invisible to AI agents, what it costs your business, and the path forward to agent-ready infrastructure.
AXIOM Scoring Framework v3.0
Scoring & Measurement
Six dimensions, quantitative scoring bands, composite grades, diagnostic failure patterns, and audit methodology.
AXIOM Build Spec v3.0
Implementation Standard
Three-tier traffic governance, axiom.json manifest, data-axiom-* markup vocabulary, readiness levels, and remediation patterns.
Open Source
AXIOM Web Reference
Open-source reference implementation, tooling, and resources for building AXIOM-compliant websites. Available on GitHub.
Related Tooling
Built for the agentic web.
See AXIOM in Action
We audited our own site. Read the case study and full audit reports.
Case Study
The Invisible Website
How one line of React code made an entire Next.js site invisible to AI agents — and the one-line fix.
Pre-AXIOM — 72/100
Initial Audit
2 Critical findings. Data Extractability at 30/100. Full six-dimension diagnostic report.
Post-AXIOM — 91/100
After AXIOM
0 Critical findings. All dimensions pass. 19-point jump from Grade C to Grade A.
Is Your Infrastructure Agent-Ready?
Most sites are invisible to the future of commerce. Find out where you stand.
Request an AXIOM Audit