geo_auditor — scoring_metrics — read-only
███████╗ ██████╗ ██████╗ ██████╗ ██╗███╗   ██╗ ██████╗ 
██╔════╝██╔════╝██╔═══██╗██╔══██╗██║████╗  ██║██╔════╝ 
███████╗██║     ██║   ██║██████╔╝██║██╔██╗ ██║██║  ███╗
╚════██║██║     ██║   ██║██╔══██╗██║██║╚██╗██║██║   ██║
███████║╚██████╗╚██████╔╝██║  ██║██║██║ ╚████║╚██████╔╝
╚══════╝ ╚═════╝ ╚═════╝ ╚═╝  ╚═╝╚═╝╚═╝  ╚═══╝ ╚═════╝
                    

OVERALL TECHFACTORS SCORE

The headline score is a weighted average of the measured weighted factors. The weighted model sums to 100 when every weighted factor is available.

If a weighted factor is intentionally not measured in a run, it is excluded from the denominator instead of being scored as zero.

Factor Weight Methodology and Signals
Authority 15 Domain age, entity schema strength, external brand signals, grounding-page coverage, and optional backlink proxy data.
Content Quality 12 Depth, formatting, and semantic coherence between the page topic and body content.
Schema Markup 10 Structured data that makes page type, entity type, and content intent more explicit.
Robots and AI Access 10 AI crawler access in robots.txt plus the quality of any existing llms.txt.
Platform Preferences Supporting Platform-specific diagnostic layer built from verified crawler/publishability signals plus lower-confidence format-fit heuristics.
HTML Structure 8 Heading hierarchy and semantic structure that help crawlers parse the page cleanly.
Freshness 5 Clear publication and update dates that reduce ambiguity around recency.
Metadata 5 Page title and description quality, clarity, and fit.
Links 5 Internal linking context plus outbound citations and authority references.
Agentic Readiness 5 WebMCP annotations and UCP discovery-file signals for agent-facing interactions.
Entities 5 Named entity extraction and resolution to canonical references such as Wikidata.
Vitals 5 Optional performance signal. When skipped for speed, it is reported as not measured and does not count against the total.
Images 4 Alt text coverage and image accessibility context.
Chunk Citability 3 How citation-ready the page's individual passages are, not just the page average.

MEASUREMENT RULES

Not every factor is available in every run. Some depend on crawl context, JS comparison, or whether a faster run intentionally skipped a slower signal.

Factor Type Current Rule
Robots, Authority, Vitals Site-wide Measured once per domain and reused across the pages in the same job.
Harmonic Centrality Supporting Only measured in crawl mode because it depends on the discovered internal link graph.
Rendering Consistency Diagnostic Measured only when static and JS-rendered HTML are both collected.
Vitals Weighted Can be intentionally skipped for speed. If skipped, it is excluded from the weighted denominator.

FACTOR STATUSES

Each factor is normalized into a product status so the UI and exports can be read quickly.

Status Meaning
strong Measured score is 80+.
acceptable Measured score is 60-79.
weak Measured score is below 60.
not_measured The signal was intentionally unavailable for that run or could not be collected under the run conditions.

A practical reading guide for the overall score: 80+ usually means strong AI readiness, 60-79 means workable but incomplete, and anything below 60 usually signals meaningful structural or trust gaps.


End of file. Click the AUDIT TERMINAL link in the sidebar to return.