{
  "id": 1913274,
  "title": "I Tested 5 AI Engines On My Own Sites. None Agreed.",
  "url": "https://urgent.news/2026/08/19/i-tested-5-ai-engines-on-my-own-sites-none-agreed",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-19T10:34:41.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/dannwaneri/i-tested-5-ai-engines-on-my-own-sites-none-agreed-4013"
  },
  "original_language": "en",
  "account": "In June, the author tested Claude and other AI engines on their own websites using SearchApi's developer program. Claude was only tested, with support for multi-model engines planned but not yet implemented. The author used SearchApi's new client, searchapi_client.py, to test ChatGPT, Gemini, Perplexity, and Bing Copilot. The code was then extended to include Claude's testing, resulting in a tool that tests five AI engines. However, Claude did not find any of the author's websites, while ChatGPT, Gemini, and Copilot found bits of information, but not all of it was accurate. The author recommends testing all AI engines to get a complete picture of AI visibility, as no single engine can predict which one will cite a website.",
  "summary": "In June I wrote that my open-source LLM visibility checker tested Claude only and that multi-model support was \"planned but not yet implemented.\" That's the kind of line that's easy to write and easy to forget. I didn't forget. I just needed a reason to finish it. The reason showed up when SearchApi launched endpoints for ChatGPT and Gemini, on top of the Perplexity and Bing Copilot endpoints…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}