Prompt optimization for TypeScript, with GEPA, SIMBA, OPRO, MIPRO, bootstrapped few-shot search, and random search behind a shared interface

Effect-native typed model programs, immutable compilation artifacts, bounded optimization, and receipted prediction.

`@elizaos/plugin-training` adds fine-tuning, trajectory management, and prompt-optimization infrastructure to an Eliza agent. Capabilities:

DSPy.ts 2.0 - Advanced declarative AI framework with multi-agent orchestration, self-learning capabilities, and state-of-the-art optimizers. Powered by AgentDB, ReasoningBank, and Swarm architecture.