Free tool · no signup
Are your docs AI-ready?
Check a sample of your docs for conflicting numbers, duplicated passages, broken anchors, and page structure. The free audit uses mechanical checks. Request a separate model-assisted answer audit to explore questions the captured sample can support. No signup.
We analyze public pages server-side and retain the dated result plus a bounded captured corpus for 90 days so share, PDF, and requested answer-audit outputs stay tied to the same evidence.
More free tools from Topicary: all free tools · llms.txt generator
How it works
We check a captured sample of your docs
We crawl a few pages
From your URL we follow the sitemap and same-section links to audit several related pages, not just the one you paste.
We cross-check them
Mechanical checks compare numbers, repeated passages, and cross-page anchors in the captured sample. These checks flag evidence to review; they do not establish factual agreement or test an assistant's answers.
We score each page's structure
Per page: heading structure, machine-readable output, answer density, structured data, and findability, all checked against the raw HTML an AI crawler actually sees.
Contradictions are an authoring problem
Reusable components keep shared passages identical across your docs. Single-sourcing reduces copy drift; authors still need to verify the facts and how readers use them.
Questions
About this tool
Do you store my content?
We analyze public pages server-side and retain the dated result plus a bounded captured corpus for 90 days so share, PDF, and requested answer-audit outputs stay tied to the same evidence.
How is this different from other “AI-readiness” checkers?
The free audit checks page structure and compares a captured sample for numeric conflicts, repeated passages, and broken anchors. Its scores are mechanical heuristics. The optional answer audit uses model-assisted checks over that same sample and reports incomplete or unavailable coverage.
Why does cross-page consistency matter for AI?
Conflicting sources can lead an assistant to contradictory or unsupported answers. Consistent source content reduces that risk, but neither single-sourcing nor this audit guarantees a correct answer.
What's in the deeper emailed report?
An answer-outcome audit plus model-assisted checks for near-duplicate conflicts, content type and self-containment, semantic terminology, and claim traceability, all using the captured sample. The report names completed, partial, and unavailable checks; it never treats a failed check as clean. Delivery is one-time; follow-up consent is separate. The on-page audit remains free.
How do you decide what counts?
We analyze the HTML captured by our crawler. Mechanical cross-page deductions cover numeric conflicts, broken anchors, and duplicate clusters; structure checks examine headings, linking, and machine-readable markup. We name the sampled pages and omissions. Partial or unavailable checks do not receive a numeric score or an all-clear claim. Model-assisted analysis belongs to the separately requested answer audit.