AI Game Testing Bot Platform

You are a QA automation engineer. Build an AI-powered game testing bot platform: Game type: [INSERT - 2D, 3D, mobile, PC] Engine: [INSERT - Unity, Unreal, custom] Bot approach: [INSERT - scripted, RL, LLM agent, hybrid] Test scope: [INSERT - crash, gameplay, UI, balance] Cover the full pipeline: 1. Bot core architecture - game state reading (memory, screen, API), action injection (input simulation, API calls), perception layer, decision layer. 2. Agent types - scripted bots for regression, exploration bots for coverage, goal-driven bots for quests, adversarial bots for multiplayer, fuzz bots for inputs. 3. Test scenarios - level walkthrough, quest completion, menu navigation, crash detection, performance monitoring, balance smoke tests. 4. Bug detection - crash and exception capture, soft-lock detection, animation/UI glitches, physics bugs, audio glitches, performance regressions. 5. Reporting - reproducible bug reports with steps, video capture, state snapshots, severity scoring, dashboard for developers. 6. Scaling - parallel bot instances, headless runners, CI/CD integration, nightly test runs, distributed execution. 7. Machine learning - train agents from playthroughs, self-improving exploration, anomaly detection, coverage optimization.