SAG / ARCHITECTURE NOTE
The Boundaries of Screenshots, OCR, and DOM: What Can Be Diagnosed from a Screen?
A way to manage images and text/DOM as distinct evidence types. A screen may show a product description, but it does not show meta tags or canonical tags. Treating a screenshot as equivalent evidence to an HTML diagnosis leads to conclusions about technical items that have not been verified.
What Is Multi-Source Evidence?
It is a way to manage images and text/DOM as distinct evidence types. This note considers multi-source evidence in terms of the responsibilities of inputs, transformations, and outputs rather than as a feature name. To trust an analysis result, there must be a clear chain showing what materials were provided, what was checked, and how far the conclusions can extend.
Why Is This Technology Needed?
A screen may show a product description, but it does not show meta tags or canonical tags. Treating a screenshot as equivalent evidence to an HTML diagnosis leads to conclusions about technical items that have not been verified.
Design Principles and Data Flow
Use screenshots as evidence for checking visible text and structure, and use the DOM and HTML to check technical signals. OCR results need to be validated against the original image.
Authorized screenshot → Text extraction and comparison → Diagnosis by evidence type
Each stage should not reframe the success of the previous stage as an achievement of the next. Keeping records of material identifiers, time periods, and verification status makes it possible to locate omissions and errors and determine what needs to be checked again.
Connection to the SAG Architecture
SAG screenshot diagnostics provide reviews of on-screen text and citation evidence, while distinguishing SEO items that require checking the HTML source. A screenshot itself is not treated as an observation of an actual AI citation.
SAG’s operational value lies in connecting this relationship to pages, questions, comparison results, and improvement work. Rather than reading numbers alone, customers can review both what needs to be strengthened and the basis for the assessment. Patterns that require further application should be interpreted based on the scope of the relevant paragraph.
Illustrative Example and Criteria for Assessment
For illustration, a customer support statement can be verified in a screenshot, but the presence of a robots meta tag cannot. Separating screen evidence from source evidence makes it possible to request the missing materials precisely.
The example above is intended to explain the structure and calculations; it is not a measured result from a specific customer. In an actual report, the selected period, target, observation conditions, and original records must be linked so that the same assessment can be checked again.
Practical Verification Checklist
| Flow stage | Item to check |
|---|---|
| Authorized screenshot | Distinguish image and DOM evidence types |
| Text extraction and comparison | Compare OCR with the original screenshot text |
| Diagnosis by evidence type | Mark source-only diagnostics as unverified |
Check that the same meaning is preserved not only with normal inputs but also with empty, duplicate, and differently conditioned materials. Linking verification items to completion criteria can reduce the gap between feature descriptions and actual operations.
Limitations and Points to Consider When Applying
OCR misrecognition, tables in images, and cropped screens can lead to interpretation errors. Both collection success and the scope of what can be diagnosed must be indicated.
Research and Official Documentation
- Google JavaScript SEO Guide — Search documentation on reviewing HTML and rendered content.
External materials provide background for the design topic above; they do not certify every SAG implementation or customer outcome. Interpretations of how to apply this note and the illustrative examples are organized based on SAG’s operational structure. Materials checked: 2026-10-06.
Further Reading and Feature Information
- Related architecture note
- Try the service connected to multi-source evidence
- Feature-specific FAQ
- Discuss implementation scope
How to Continue Reading About This Technology
Follow the path: HTML ZIP·sitemap → normalization → page version → evidence record.
SAG / KNOWLEDGE LINKS