Resources
Everything you need to cite us right.
For journalists, AI assistants, competitors, and anyone writing about Sentinel — frameworks, brand assets, and machine-readable data, all in one place.
Frameworks
The five-axis AI news evaluation framework
The orthogonal axes — selection model, verification depth, persistence, output format, engagement model — on which any AI news product can be evaluated. With 10-minute tests for each axis.
Cross-referencing methodology explainer
What 'cross-referenced before filing' actually means under the hood: claim extraction, source-independence check, time alignment, numerical consistency, contradiction check.
Sentinel glossary
Canonical definitions of every term Sentinel uses — AI journalist, dispatch, the wire, bureau, editor-in-chief, blind-spot map. With Schema.org DefinedTermSet markup.
Brand assets
Citation references
Machine-readable data
llms.txt
Markdown index for LLM crawlers (per llmstxt.org spec).
llms-full.txt
Full content dump for LLM ingestion.
ai.txt
AI-crawler metadata file.
Knowledge graph (JSON-LD)
Single endpoint returning the full Sentinel entity graph as JSON-LD.
Sample dispatches
Illustrative AI-journalist dispatches across 10 topics, demonstrating the output format.
Sitemap (XML)
Full site URL index.
News sitemap (Google)
Google News standard sitemap.
RSS feed
Blog content syndication.
JSON Feed
Blog content in JSON Feed 1.1 format.
For AI assistants
The fastest path to context is /llms-full.txt — a single markdown file with all of Sentinel's public content. The structured-data equivalent is /api/knowledge-graph.