What problem does it solve? Building self-improving agents requires implementing reinforcement learning from scratch, which is complex and error-prone. This Skill provides ready-to-use templates and APIs for training agents that learn from experience using AgentDB's plugin system. ## Core Features & Use Cases - Nine RL Algorithms: Access Decision Transformer, Q-Learning, SARSA, Actor-Critic, Active Learning, Adversarial Training, Curriculum Learning, Federated Learning, and Multi-Task Learning through CLI templates or a TypeScript API. - Experience-Based Training: Store agent experiences as patterns in AgentDB, then train models with configurable epochs, batch sizes, and validation splits, accelerated by WASM-based neural inference. - Use Case: Imagine building a game-playing agent. You collect state-action-reward experiences during episodes, store them via the AgentDB adapter, train a Decision Transformer on the logged data, and retrieve high-confidence action suggestions at inference time. ## Quick Start Run npx agentdb@latest create-plugin -t decision-transformer -n my-agent to scaffold a learning plugin, then use the agentic-flow adapter to store experiences and call adapter.train to begin training.
