thedotmack/claude-mem
Claude-Mem is a plugin for Claude Code that captures, compresses, and preserves coding session context to enable persistent memory and continuity across sessions.
Awesome AI Agents › Continual Learning Agents
AgentDock/AgentDock
AgentDock is an advanced framework designed for building sophisticated AI agents capable of performing complex tasks with configurable determinism. It consists of two main components: AgentDock Core, an open-source backend-first framework that is framework-agnostic and provider-independent, allowing developers full control over their AI agents' implementation; and an Open Source Client, a complete Next.js application that serves as a reference implementation and consumer of the core framework. The framework is built with TypeScript and emphasizes simplicity, extensibility, and configurable determinism, making it ideal for creating reliable and predictable AI systems that require minimal supervision. AgentDock supports a variety of AI agents, including specialized examples like Dr. Gregory House, a diagnostic reasoning agent that uses multi-stage workflows and tools such as search, deep research, and PubMed to solve complex medical cases; Cognitive Reasoner, a multi-stage reasoning engine that orchestrates seven specialized cognitive tools to solve complex problems with human-like reasoning; History Mentor, an educational agent that combines vectorized historical knowledge with search capabilities and dynamic diagram rendering to create immersive learning experiences; and Calorie Vision, a vision-based nutritional analysis system that uses computer vision and structured data extraction to provide detailed nutrient breakdowns from food images. The framework is built on core design principles such as simplicity first, node-based architecture where all capabilities are implemented as nodes, tools as specialized nodes extending agent capabilities, configurable determinism to balance creativity and predictability, and comprehensive type safety through TypeScript. Configurable determinism allows developers to control the level of predictability in agent behavior by defining deterministic workflows or allowing non-deterministic LLM-based reasoning where needed. This balance enables both creative AI outputs and reliable system behavior. AgentDock supports deterministic workflows typical of workflow builders, ensuring predictable execution paths and outcomes, while also enabling non-deterministic agent behavior through LLM inference for adaptability. It uniquely combines non-deterministic agent intelligence with deterministic sub-workflows, offering the best of both worlds for AI application development. The project is actively maintained, open-source under the MIT license, and provides extensive documentation and community support via Discord and other channels.
https://github.com/AgentDock/AgentDock
Also in Continual Learning Agents
vectorize-io/hindsight
Hindsight is an agent memory system designed to create smarter, learning AI agents, addressing limitations of traditional memory techniques like RAG and knowledge graphs.
langchain-ai/langchainjs
LangChain.js is a TypeScript framework for building context-aware, reasoning applications powered by large language models, offering modular components, chains, agents, and tools for production and deployment.
aden-hive/hive
Hive is an outcome-driven agent development framework designed for evolving agents with AI evaluation, human-in-the-loop, and self-improvement capabilities.
aiming-lab/MetaClaw
Proxy-based layer that turns everyday conversations with a personal agent into learning signals, injecting skills per turn and running scheduled reinforcement learning updates without a local GPU.
kayba-ai/agentic-context-engine
Agentic Context Engine (ACE) is a framework that enables AI agents to autonomously learn from their successes and failures, continuously improving their performance across various tasks without additional training data or fine-tuning.
modelscope/AgentEvolver
Self-evolving agent training framework from ModelScope that combines automatic task generation, experience-guided exploration and attribution-based credit assignment into one reinforcement learning stack.
ghostwright/phantom
Self-hosted persistent AI co-worker built on the Claude Agent SDK that runs on its own machine with long-term memory, self-evolution, an MCP server and Slack, Telegram, email and web channels.