This is the index for the free tools. Each runs locally in the browser, and each has its own page explaining what it does.
01What this section covers.
2 groups: the tools themselves, and the pages that document how the JSON-LD validator processes, sources and secures what it reads.
02The pages.
The JSON-LD Schema Validator, what valid means
A standalone public product that tests structured data in code, HTML and live pages against JSON, JSON-LD, Schema.org and consumer requirements.
llms.txt: a curated map of your website for AI systems
llms.txt is a Markdown file at the root of a website: a short description and a curated list of its most useful pages, written for AI systems.
agent.json: telling AI agents what your service does
The JSON file published at a fixed address on your domain that tells AI agents what your service can do, where its endpoint is and how to authenticate.
/ask endpoint: answering questions in plain language
/ask is an address on your site that takes a question in plain language and answers from your own content, defined by NLWeb, with 4 routes to build one.
Cloudflare Markdown for Agents, explained
A Cloudflare feature that converts a page from HTML to Markdown at the network edge when an AI agent asks. What it does, who owns it, and how to enable it.
Akamai: serving pages as Markdown to AI agents
Serving a web page as clean Markdown when an AI agent asks for it. The Akamai position on AI traffic, 3 routes to build it, and Bot Manager policy.
Working with the JSON-LD Schema Validator
Submission to result, source-linked findings, graph navigation, the states a check can return, a specimen finding, and versioned export.
The validator processing sequence, stage by stage
The processing stages, from intake to result assembly, with the evidence each stage accepts, the checks it applies, and the output it assures.
Validator primary sources, each finding traced
The published standards the validator treats as authority, from JSON and JSON-LD to Schema.org and consumer documentation, each with its role.
Validator security and privacy, data handling
What happens to what you submit: security architecture, privacy and isolation, data handling, release assurance, and the recorded behaviour per evidence case.
03Where to start.
The JSON-LD validator is the most used of these. Its processing, sources, security and privacy are each documented separately.
Next
Read what AI reads about you, then decide.