What problem does it solve? AI agents typically repeat the same mistakes and cannot improve from past experience. This Skill implements ReasoningBank's adaptive learning system so agents can record task outcomes, recognize recurring patterns, and select better strategies over time. ## Core Features & Use Cases - Experience Recording & Strategy Recommendation: Log task outcomes with context and metrics, then query the optimal strategy for a given task type and situation. - Pattern Recognition & Meta-Learning: Learn patterns from data, match them against current situations, and perform meta-learning and transfer learning across domains. - Persistent Storage with AgentDB: Store learned patterns in AgentDB with vector search for semantic pattern matching and confidence-filtered queries. - Use Case: Build a self-improving code review agent that records each review's outcome, compares strategies like TDD-first versus debug-first, and automatically recommends the highest-scoring approach for the next task. ## Quick Start Ask the agent to initialize ReasoningBank with AgentDB persistence, record a task outcome, and recommend the best strategy for your next task.
