28QualityDraftQuality: 28/100LLM-assigned rating of overall page quality, considering depth, accuracy, and completeness.12.5ImportancePeripheralImportance: 12.5/100How central this topic is to AI safety. Higher scores mean greater relevance to understanding or mitigating AI risk.15ResearchMinimalResearch Value: 15/100How much value deeper investigation of this topic could yield. Higher scores indicate under-explored topics with high insight potential.
Content2/13
SummarySummaryBasic text summary used in search results, entity link tooltips, info boxes, and related page cards.ScheduleScheduleHow often the page should be refreshed. Drives the overdue tracking system.Set updateFrequency in frontmatterEntityEntityYAML entity definition with type, description, and related entries.Add entity YAML in data/entities/Edit historyEdit historyTracked changes from improve pipeline runs and manual edits.crux edit-log view <id>OverviewOverviewA ## Overview heading section that orients readers. Helps with search and AI summaries.Add a ## Overview section at the top of the page
Tables1/ ~1TablesData tables for structured comparisons and reference material.Diagrams0DiagramsVisual content — Mermaid diagrams, charts, or Squiggle estimate models.Add Mermaid diagrams or Squiggle modelsInt. links0/ ~3Int. linksLinks to other wiki pages. More internal links = better graph connectivity.Add links to other wiki pagesExt. links0/ ~2Ext. linksLinks to external websites, papers, and resources outside the wiki.Add links to external sourcesFootnotes0/ ~2FootnotesFootnote citations [^N] with source references at the bottom of the page.Add [^N] footnote citationsReferences0/ ~1ReferencesCurated external resources linked via <R> components or cited_by in YAML.Add <R> resource linksQuotes0QuotesSupporting quotes extracted from cited sources to back up page claims.crux citations extract-quotes <id>Accuracy0AccuracyCitations verified against their sources for factual accuracy.crux citations verify <id>RatingsN:2 R:3 A:4 C:5RatingsSub-quality ratings: Novelty, Rigor, Actionability, Completeness (0-10 scale).
The Quantitative Claims tool scans all MDX content for numbers, percentages, dollar amounts, and other quantitative data that could be extracted as standalone insights.
Patterns Detected
Type
Examples
Description
Percentages
40%, 30-50%
Success rates, probabilities, coverage
Dollar Amounts
$1B, $10 million
Research funding, market sizes, costs
People Counts
500 researchers
Team sizes, community estimates
Timelines
by 2030, in 5 years
Predictions, forecasts
Multipliers
10x, 3-fold
Performance gains, risk increases
Probabilities
20% probability
Risk estimates, likelihood
Large Numbers
100 billion
Scale metrics
Notable Claims
Claims are marked as "Notable" when their surrounding context contains importance indicators like:
catastrophic, existential, critical
surprising, unexpected, contrary
only, merely, just (indicating scarcity)
most, majority (indicating prevalence)
unprecedented, first
Generating the Data
The claims data is generated by running:
cd apps/longterm
node scripts/find-quantitative-claims.mjs
This scans all MDX files and outputs to src/data/generated/quantitative-claims.json.
Re-run this script periodically to capture new content.
Using the Tool
Filter by type to focus on specific claim types (percentages, dollars, etc.)
Toggle "Only notable" to see claims with importance indicators
Search for specific topics or numbers
Click through to source pages to verify context
Extract as insight if the claim is surprising, important, and well-sourced