Alpha This system is experimental. Scores and classifications are early-stage research and may be unreliable. Methodology → home / data Dataset & Exports
The HRO evaluation dataset — story scores, domain profiles, and rater evaluations — will be available for bulk download here. Dataset exports and bulk snapshots are in progress.
API Access (available now)
Endpoint Description /api/v1/stories Paginated evaluated story feed with HRCB scores /api/v1/story/{id} Single story with per-model rater evaluations /api/v1/articles Per-UDHR-provision aggregate scores: editorial/structural averages, story counts, trigger counts, evidence distribution, SETL tension /api/v1/signals Corpus-wide signal aggregates: transparency disclosure rates, accessibility metrics, temporal framing, tone/scope distributions. Includes generated_at timestamp for build-time consumers. /api/v1/domains Domain signal aggregates (HRCB, EQ, SO, TD, valence, arousal) /api/v1/domain/{domain} Domain profile with recent evaluated stories /api/v1/domain/{domain}/history Daily HRCB profile snapshots (?days=30, max 365) /api/v0/topstories.json HN-compatible: top 500 story IDs by HN score /api/v0/beststories.json HN-compatible: top 500 story IDs by HRCB score /api/v0/item/{id}.json HN-compatible item + hcb extension object
All endpoints are public, CORS-enabled, and rate-limited to 200 requests/hour per IP.
OpenAPI 3.1 spec ·
Methodology reference →
Bulk Exports (coming soon)
File Format Contents /api/v1/export/stories.csv CSV All evaluated stories with full signal data /api/v1/export/stories.jsonl JSONL Same as CSV, newline-delimited JSON /api/v1/export/domains.csv CSV All domain signal profiles /api/v1/export/rater-evals.jsonl JSONL Per-model evaluation records (ML research)
License and citation information will be published here when bulk exports are available.