AI-native ranking engine. The dynamic top 11 in every niche — 10 ranked plus one wildcard — refreshed continuously, built for AI agents to read, query, and cite.
🌐 Live site: https://topelevens.com
📖 Methodology: https://topelevens.com/methodology
🤖 Agents: https://topelevens.com/agents.json · https://topelevens.com/llms-full.txt · https://topelevens.com/openapi.json · https://topelevens.com/mcp
📡 API: https://topelevens.com/api/lists/{slug} → JSON · .csv · .md
The web is full of ranking sites that are either:
- Pay-to-play (G2, Capterra) where vendors buy placement,
- Stale (Best-of lists updated once a year if at all), or
- Hand-written for humans only (no machine-readable surfaces).
Top 11 is none of those. Autonomous AI curators score providers against a public, weighted methodology, rankings refresh continuously, and the whole site is built so AI agents can consume it without a human intermediary.
app/— Next.js 15 App Router sourcedata/— Per-list JSON data (the source of truth for rankings)lib/— Schema generation, list loading, scoring helpersscripts/— Build helpers (build-llms-full.mjs,indexnow-push.mjs)docs/seo-aeo-master.md— Living SEO + AEO + GEO playbook
| Slug | Vertical |
|---|---|
/fractional-cfo |
Finance |
/dental-crm |
Healthcare |
| Surface | Purpose |
|---|---|
/llms.txt |
Concise overview for LLMs |
/llms-full.txt |
Full machine-readable corpus |
/agents.json |
Agent capabilities manifest |
/openapi.json |
OpenAPI spec for the public API |
/mcp |
Model Context Protocol endpoint |
/.well-known/ai-plugin.json |
ChatGPT Action manifest |
/api/lists/{slug} |
Per-list JSON / CSV / Markdown |
npm install
npm run dev # http://localhost:3000
npm run build # also rebuilds /llms-full.txtSpotted a ranking error? Open an issue with the list URL, the entry rank, the specific methodology criterion you believe was misapplied, and evidence. Credible complaints land in the public changelog on the affected list.
Source: MIT. Editorial commentary, scoring formulas, and ranking outputs: © Top 11.