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.
Awesome AI Agents › Continual Learning Agents
thedotmack/claude-mem
Claude-Mem is a persistent memory compression system designed specifically for Claude Code, an AI coding assistant. This plugin automatically captures all activities and interactions Claude performs during coding sessions. It then uses AI, leveraging Claude's agent-sdk, to compress and summarize this information semantically. The compressed memory is stored and injected back into future sessions, allowing Claude to maintain continuity and context across different coding sessions without losing track of previous work or project history. The system works by capturing tool usage observations such as reading, writing, and other interactions during a session. These observations are processed by a worker service that extracts meaningful learnings and compresses them into semantic summaries. When a new session starts, recent observations are injected as context, enabling Claude to recall relevant past information automatically. Key features include persistent memory that survives across sessions, progressive disclosure which layers memory retrieval with token cost visibility, and skill-based search that allows querying project history efficiently. It also offers a web viewer UI for real-time memory streaming, a Claude Desktop skill for searching memory from conversations, and privacy controls to exclude sensitive content from storage using special tags. The plugin requires minimal manual intervention, operating automatically once installed. It supports fine-grained context configuration and provides citations for past observations, accessible via a web interface. Additionally, it offers a beta channel for experimental features like Endless Mode. Claude-Mem is well-documented with comprehensive guides on installation, usage, architecture, configuration, and troubleshooting. It is licensed under AGPL 3.0 and integrates seamlessly with Claude Code, enhancing the AI's ability to remember and utilize past coding session data effectively, improving productivity and continuity for developers.
https://github.com/thedotmack/claude-mem
Also in Continual Learning Agents
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.
AgentDock/AgentDock
AgentDock is an open-source framework for building sophisticated AI agents with configurable determinism, enabling reliable and creative AI applications through a node-based architecture and multi-stage workflows.
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.