DEV Community

Fey
Fey

Posted on • Edited on

What is XanLens? The GEO Tool Built for AI Agents

#seo #ai #webdev #startup

TL;DR

XanLens is a Generative Engine Optimization (GEO) API that audits your brand's visibility across AI search engines, then generates content to fix it. Built for AI agents, not dashboards.

  • Audit: Free. Checks how Gemini, ChatGPT, and Perplexity talk about your brand.
  • Fix: $0.49/call via USDC. Generates blog posts, FAQ schema, llms.txt, social posts, about page copy, and optimization plans.

The Problem

SEO is dead for AI search. When someone asks ChatGPT "what's the best tool for X?", your Google ranking doesn't matter. What matters is whether the AI knows you exist.

Most brands score poorly. We audited ourselves and got 40/100. That's the honest truth.

How It Works

  1. Query AI engines with branded + discovery prompts
  2. Analyze mentions, sentiment, citations
  3. Score visibility 0-100 with letter grade
  4. Generate 7 types of optimized content to improve the score

What the Fix Generates

  • Blog post optimized for AI citation patterns
  • FAQ schema (JSON-LD) for structured data
  • Social media posts for platform-specific visibility
  • About page copy with entity-friendly language
  • Schema markup recommendations
  • llms.txt file for AI crawler discovery
  • robots.txt audit for AI crawler access

Why Agents, Not Humans?

Human SaaS tools need dashboards, onboarding flows, customer support. We skipped all that.

XanLens is a REST API. AI agents call it, pay with USDC via x402, get results. No API keys, no accounts, no subscriptions.

# Free audit
curl -X POST https://xanlens.com/api/v1/audit \
  -H "Content-Type: application/json" \
  -d '{"brand": "your-brand", "industry": "your-industry", "url": "https://your-site.com"}'
Enter fullscreen mode Exit fullscreen mode

Integrations

  • REST API — Any HTTP client
  • MCP Server — For Claude, OpenClaw, and agent frameworks
  • OpenClaw Skill — xanlens-audit@2.0.0 on ClawHub

What We Learned Building This

  1. Llama hallucinates about unknown brands — it gave us 100/100 by making up features we don't have. We removed it from the engine.
  2. Web presence matters more than on-site SEO — Reddit, LinkedIn, GitHub, Dev.to articles are what AI engines actually cite.
  3. Schema markup is table stakes — Organization + FAQPage JSON-LD are minimum requirements.
  4. llms.txt is the new robots.txt — AI crawlers look for it. Most sites don't have one.

Try It

Built by @DeCryptobean and Xan (AI partner).

This is part of our building-in-public series. We're eating our own dogfood — applying XanLens recommendations to XanLens itself and tracking the score over time.

Top comments (0)