Everything in the package is static and uploadable. The files help AI systems discover, understand, and accurately describe your business without requiring code changes or a new backend. The exact package is driven by standards and your site's profile, not by a fixed file count. Typical non-commerce sites receive 12 files after the current standards-conformance update.
Google says its generative AI features are rooted in core Search systems. For Google, the fundamentals still matter most: crawlability, indexability, helpful content, clear technical structure, page experience, and standard SEO eligibility.
Files like `llms.txt` are not required for Google Search AI features. We include them for the broader AI and agent ecosystem, while your installation instructions also include a Google AI Search fundamentals checklist.
Separate from Google Search, Chrome's experimental Lighthouse Agentic Browsing audits now check for an `llms.txt` file at the domain root. The MVP package includes `llms.txt`, but we do not claim to pass the full Agentic Browsing audit set.
Google Chrome's experimental Lighthouse Agentic Browsing category checks four things: `llms.txt` presence, WebMCP integration, accessibility tree integrity, and layout stability.
The MVP package includes the root-level `llms.txt` file Chrome's audit looks for.
We include an experimental static `webmcp-registration.js` wrapper, but do not count WebMCP toward Lighthouse coverage until the audit path is verified.
Accessibility tree integrity depends on your site's existing HTML, labels, and interactive elements.
Cumulative Layout Shift depends on your site's CSS, scripts, images, fonts, and rendering behavior.
A lightweight instruction file placed at your website root. It gives AI tools and agents that choose to read it a clear overview of your site and how they should use the rest of the package.
Business value: reduces guessing for compatible AI tools. It is featured in Chrome's experimental Lighthouse Agentic Browsing audits, but is not required for Google Search AI features.
Your standard robots file updated with modern AI signals and references to the agent-readiness package.
Business value: keeps traditional crawler guidance and AI-era signals aligned in one familiar standards-backed place.
A machine-readable permissions file placed in the `.well-known` folder. It defines safe boundaries for agent use.
Business value: gives agents structured guidance instead of forcing them to infer what is allowed.
A strategic markdown overview of your business, services, capabilities, and recommended agent reading order.
Business value: gives AI systems a more complete brand narrative than they can get from page titles alone.
Validated Schema.org structured data generated from your actual website content and merged with existing JSON-LD where appropriate.
Business value: supports structured data quality and rich-result eligibility. It is useful SEO infrastructure, not special Google AI markup.
Our richer implementation of the AI Info / LLM Info Page convention originated by Amin Foroutan in April 2025 and amplified by Steve Toth in June 2025. It includes extracted services, pricing language, testimonials, FAQs, use cases, example prompts, a Company Snapshot JSON block, version, and contact guidance.
Business value: creates one authoritative page humans and AI systems can use instead of piecing together information from scattered pages.
An A2A protocol card placed in `.well-known` for static agent-to-agent discovery metadata.
Business value: gives agents a structured capability card rather than forcing them to infer what the business can do.
A static catalog of skills and capabilities that agents can read to understand how to describe the business.
A static Model Context Protocol server card describing agent-readable capabilities and resources. It prepares the site for agent discovery without claiming a live MCP server is installed.
An experimental static WebMCP registration wrapper that describes discoverable actions and context for agents. We ship the file, but do not yet claim it satisfies Chrome's Lighthouse WebMCP audit until the audit's expected install pattern is verified.
A standards-backed API catalog path using IETF Linkset format. For customers without a public REST API, we emit a valid empty placeholder that clearly says no public API is advertised.
A Universal Commerce Protocol profile. For non-commerce customers, we emit a minimal no-commerce profile; server-side commerce protocols such as x402, AP2, and ACP remain roadmap items.
Earlier AgentReadyWebsite drafts described the package as 14 files because that was what the generator produced. The standards audit changed the better answer: ship the files with published authority, and stop counting non-standard extras as a feature.
`agentic-robots.txt` is folded into standard `robots.txt`, `nlweb-config.json` is removed because NLWeb is a deployable runtime rather than a static file, and `.well-known/commerce-manifest.json` is renamed to `.well-known/ucp` to align with Google's Universal Commerce Protocol.
No service can honestly promise that `llms.txt` or AI-specific files will make you rank in Google AI Overviews or AI Mode. Google has been clear that those features do not require `llms.txt` or other special AI-only markup.
AgentReadyWebsite focuses on a different surface: agentic browsing. When AI assistants, browser agents, research agents, or coding agents land on your domain, the package gives them cleaner context and clearer boundaries.
Both things can be true: Google Search does not require these files for AI Search placement, while Chrome's developer tooling now recognizes `llms.txt` as part of experimental agent-readiness auditing.
AgentReadyWebsite delivers static files and installation guidance. It does not install plugins, create live server-side MCP/A2A/NLWeb/WebMCP endpoints, host your website, rewrite your site's accessibility tree or layout stability, or guarantee traffic and revenue lift.