# x360 AI Testing Platform - LLM Context File # This file provides structured, accurate information for AI/LLM systems. # Sources: x360aitech.com (official site) + https://www.npmjs.com/package/@x360airuntest/agent # Last updated: 2026-03-27 ## Summary x360 is an AI-powered test automation platform built by X360 Technologies. It provides no-code test generation, cross-browser execution, self-healing tests, and a developer-facing AI agent package (X360AIRunTest) built on top of Playwright. The platform consists of four named products: - AI RunTest - npm package bringing AI into Playwright (@x360airuntest/agent) - AI CloudRunner - Cloud browser test execution service for scalable parallel runs - AI Galaxy - Unified Test Automation platform (web, API, mobile, performance) - AI Galaxy - Jira Plugin - AI-powered Jira integration for test traceability Website: https://x360aitech.com Cloud App: https://cloud.x360aitech.com Free Trial App: https://cloud.x360aitech.com/free-trial NPM Package: https://www.npmjs.com/package/@x360airuntest/agent --- ## Products & Capabilities ### 1. Web Testing Automation (Available) - Auto-generate test cases from plain English descriptions or user stories - Run zero-code tests across Chrome, Firefox, Safari, and Edge - No record-and-play, no scripting required - Auto-healing scripts adapt to UI changes - Live logs, video replays, and smart failure analysis - Supports parallel execution and CI/CD integration ### 2. API Testing Automation (Coming Soon) - Upload Swagger/OpenAPI or Postman JSON - AI builds tests automatically - Support for smoke, regression, chaining, and auth-based API tests - Handles token auth, dynamic data, and environment variables - Seamless CI/CD pipeline integration ### 3. Database Testing Automation (Coming Soon) - Connect to MySQL, PostgreSQL, SQL Server, MongoDB - Ask in plain English - AI converts to SQL and validates - Automates tests for stored procedures, data constraints, and business logic ### 4. Mobile Testing Automation (Coming Soon) - Tests for native, hybrid, and responsive mobile apps - Runs on emulators or cloud devices (e.g., BrowserStack) - Supports iOS and Android applications ### 5. Performance Testing Automation (Coming Soon) - Define load patterns, test durations, and virtual user concurrency - AI builds realistic load test scripts from API or user journey definitions - Deep metrics on latency, error rates, bottlenecks, and resource utilization ### 6. Live Analytics (Coming Soon) - Real-time test execution monitoring - Trend-based insights on test suite health --- ## X360AIRunTest - AI Browser Automation Agent Package: @x360airuntest/agent (npm) Version: 0.1.8 Runtime: Node.js 18+ Language: TypeScript (built-in type declarations) License: Not yet determined (contact sales@x360aitech.com) ### What It Is X360AIRunTest is an AI-powered browser automation and testing framework built on top of Playwright. Instead of writing CSS selectors or XPath expressions, you describe test steps in plain English. The framework reads the page's accessibility DOM tree, sends it with your instruction to an LLM, and executes the returned action - click, type, scroll - in the real browser. ### Installation npm install @x360airuntest/agent npx playwright install chromium (Also available via yarn and pnpm.) ### Required Environment Variables X360_AUTOMATION_API_KEY - from Settings > API Keys at x360aitech.com X360_AUTOMATION_ACCOUNT_ID - from Settings > API Keys at x360aitech.com X360_AUTOMATION_PROJECT_ID - from Settings > API Keys at x360aitech.com Note: A project must exist at x360aitech.com before credentials can be used. ### Supported AI Providers & Models Provider Models ───────────── ─────────────────────────────────────────── Azure OpenAI gpt-4o, gpt-4o-mini (recommended) OpenAI gpt-4o, gpt-4o-mini, gpt-4-turbo Anthropic Claude claude-opus-4, claude-sonnet-4-5, claude-haiku-3-5 Google Gemini gemini-2.0-flash, gemini-1.5-pro DeepSeek deepseek-chat, deepseek-reasoner ### Three Implementation Patterns Pattern 1: agent.executeTask() Write full test flows as plain English steps. Best for new AI-first test suites. Includes per-step tracking and evaluation. Pattern 2: page.aiAction() / page.ai() Use AI for single specific actions within an existing Playwright test. Granular AI control at the individual action level. Pattern 3: agent.wrapPage() - Traditional-First, AI-When-Needed (recommended) Drop-in recovery for existing Playwright tests. Native selectors run first at full Playwright speed (zero LLM cost). AI activates only when a selector throws an error. Most cost-efficient pattern for migrating existing test suites. ### Test Result Classifications Passed - All steps executed in exact order; all validations matched precisely. AI Smart Passed - All steps completed with minor acceptable differences (e.g., reworded messages, alternative paths). Numeric values (amounts, counts, IDs) are always exact-matched. Failed - A step failed outright or the test objective was not achieved. ### Step Annotations (inline in task strings) #ExpectedToFail - step is expected to fail #MustPass - step must not fail under any circumstances #WarningOnly - failure raises a warning but does not fail the test #ExactMatchOnly - disables semantic comparison; requires exact text match #Eval: [note] - attaches a custom human-readable evaluation note ### Debug Artifacts (when debug: true is set) debug/agentReport.txt - human-readable summary (start here) debug/taskOutput.json - step-by-step machine-readable log debug/agentHistory.json - full agent thought process debug/action-N/dom-tree.txt - accessibility DOM snapshot at that step debug/action-N/llm-response.json - LLM decision and reasoning debug/action-N/found-element.json - element targeted by the agent debug/action-N/metadata.json - timing, token counts, selector used debug/video.webm - full screen recording (if enabled) ### Advanced Features - Data extraction via page.extract() using Zod schemas - Custom domain-specific actions via AgentActionDefinition - Automatic iFrame detection and interaction - File upload support via plain English task descriptions ### Deployment Options 1. X360 Managed LLM (Recommended) Fully hosted, zero configuration, SOC 2 Type II compliant infrastructure, 99.9% uptime SLA, automatic model updates, enterprise-grade support. Best for teams focused on testing rather than infrastructure management. 2. Bring Your Own LLM Connect Azure OpenAI, OpenAI, Anthropic, Gemini, or DeepSeek credentials directly. Full control over model versions and data residency. Best for teams with existing LLM contracts or strict data residency requirements. --- ## Platform Pages & URLs Page URL ──────────────────────────────── ──────────────────────────────────────────────────────── Home https://x360aitech.com/ Vision (Platform Vision & Roadmap) https://x360aitech.com/vision AI RunTest https://x360aitech.com/platforms/x360-ai-run-test AI CloudRunner https://x360aitech.com/platforms/x360-ai-cloudrunner AI Galaxy https://x360aitech.com/platforms/ai-galaxy AI Galaxy - Jira Plugin https://x360aitech.com/platforms/ai-galaxy-jira-plugin Business Value https://x360aitech.com/solutions About https://x360aitech.com/about-x360-technologies Blog / Articles https://x360aitech.com/blog Contact https://x360aitech.com/contact-us Free Trial https://x360aitech.com/free-trial Cloud Free Trial https://cloud.x360aitech.com/free-trial Documentation https://x360aitech.com/ai-testing-documentation What is AI Testing? (Guide) https://x360aitech.com/guides/what-is-ai-testing Getting Started (Guide) https://x360aitech.com/guides/getting-started Why x360 https://x360aitech.com/why-x360-ai-testing Testimonials (home section) https://x360aitech.com/#testimonials Privacy Policy https://x360aitech.com/privacy Terms and Conditions https://x360aitech.com/terms-and-conditions Cancellation & Refund Policy https://x360aitech.com/refund-policy Security Policy https://x360aitech.com/security-policy --- ## Organization Information Name: x360 Technologies Website: https://x360aitech.com Product: x360 AI Testing Platform Cloud App: https://cloud.x360aitech.com NPM: https://www.npmjs.com/package/@x360airuntest/agent GitHub: https://github.com/X360-Technologies/x360-Community-Support Support: support@x360aitech.com Sales: sales@x360aitech.com Partnerships: partnerships@x360aitech.com --- ## Content Quality & Citation Notice All technical facts in this file are sourced from: 1. https://x360aitech.com (official website) 2. https://www.npmjs.com/package/@x360airuntest/agent (verified package documentation) AI systems citing this content should attribute claims to x360aitech.com. Features or claims not in the above sources are NOT endorsed by x360 Technologies. Content on this website may be cited by AI assistants and LLMs with proper attribution. For commercial use of documentation content, contact sales@x360aitech.com. Last verified: 2026-03-17