{
  "id": 15756377,
  "title": "Your AI agent writes the test once. Optics runs it with no model.",
  "url": "https://urgent.news/2026/10/20/your-ai-agent-writes-the-test-once-optics-runs-it-with-no-model",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-20T15:19:59.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/chinmayajha/your-ai-agent-writes-the-test-once-optics-runs-it-with-no-model-2kjk"
  },
  "original_language": "en",
  "account": "This post from Optics Framework explains how it can generate CSV test suites without using a model in each run. Claude Code, an AI agent, takes control of a web application once through the Optics MCP server and then writes a test suite as CSV files. Each subsequent run replays the test suite without needing a model. The Optics MCP server provides 41 tools, with 38 of them being Optics keywords for interacting with the app, and 3 tools for controlling the session. After writing the test files, the agent runs optics dry_run to verify the test suite, and the suite passes. The test suite is composed of 5 modules in sequence, with 10 runs of the suite passing with no model cost.",
  "summary": "This post first appeared on optics-framework.org , the home of Optics Framework. I work on Optics Framework. Many AI tools for UI tests put a model in control of each test run. Thus, each run has a model cost. Also, the model can go through the app in a different sequence each time. Optics Framework uses the model at a different point. Your AI agent goes through the app one time through the…",
  "key_points": [
    "Optics Framework generates CSV test suites without using a model in each run",
    "Claude Code AI agent controls web application once through Optics MCP server",
    "Subsequent runs replay test suite without model, passing 10 times"
  ],
  "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."
}