Report card

Do AI agents pick Starlight?

How often agents choose Starlight when a developer needs documentation platforms — measured across Claude, GPT, Gemini.

Starlight

0%Pick Rate
95% CI 0%–2%

When developers ask an AI agent for docs platforms, Starlight is picked 0% of the time — ranking #7 of 9 measured across Claude, GPT, Gemini.

Picked 0% Named, not picked 5% Never surfaced 94%

Beaten by Redocly (9%), Docusaurus (8%), Nextra (7%), Scalar (5%), Mintlify (4%), ReadMe (4%).

Awareness

0 docs in FineWeb (10BT sample) · 174 URLs in Common Crawl

In the raw web crawl, but thin in the filtered corpus models actually train on — an awareness gap at the source.

Agent readiness

Level 0 · Not Ready · 1/16 agent standards

Starlight is missing Link headers, DNS-AID records, Markdown for agents, AI bot rules. Readiness is what a site publishes for agents, not whether agents pick it — across our corpus the two barely correlate. Scored by Cloudflare's Agent Readiness score.

Get the full report

The per-model and per-surface breakdown for Starlight — where it wins, where it loses, and to whom — plus an alert when the Pick Rate moves.

About Starlight

A documentation site builder built on top of the Astro framework.

Starlight is a documentation website framework that runs on Astro, designed for developers and teams who want to publish technical docs. It handles the structure, navigation, theming, and performance concerns that come with documentation sites, so you can focus on writing content. It targets teams who want a fast, well-structured docs site without building the scaffolding from scratch.