Structure & Semantics
Review document structure, landmarks and semantic signals in the rendered page.
AXIA Audit
An independent evaluation tool for web experience quality, combining HTTP evidence, controlled Chromium sessions, rendered DOM, accessibility-tree data and targeted probes.
Evaluation scope
Explore the evaluation areas. Coverage depends on the inspected page, configured profiles and evidence available during a run.
Review document structure, landmarks and semantic signals in the rendered page.
Inspect titles, metadata and indexing directives with their intended context.
Combine tested accessibility evidence with criteria that require human review.
Inspect configured desktop, tablet and mobile profiles for layout and overflow evidence.
Collect observed loading, interaction and layout-stability measurements within the tested session.
Inspect captured requests, failed resources and HTTP response evidence.
Surface captured exceptions, promise rejections and console messages for interpretation.
Inspect browser safety signals, secure context and mixed content within the observed scope.
Interpretation
New Runs use WCAG 2.2 A/AA as their primary conformance scope, retaining AAA technical results outside primary counts. Automated evaluation does not establish complete criterion coverage or certification.
A stored Run’s policy snapshot determines interpretation. Historical Runs are not re-audited by the current engine.
Automated checks collect tested evidence. Assisted criteria can request human review. Observations provide context without an independent pass/fail status.
No aggregate AXIA Audit Score or AAA certification is implemented. Browser safety evaluation is not a penetration test.
Results & evidence
Evaluation boundaries
Viewports and interactions are sampled. Some measurements may be unavailable or inconclusive in the headless runtime. An unavailable measurement is not a pass.
Optional SEO resources can be advisory. Human decisions are stored separately. Optional AI-assisted reports summarize supplied results; they do not execute additional tests.