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
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.
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.
ray-r-ren/agent-apprenticeship
Ecosystem and CLI where local coding agents complete real tasks in workflow loops, are evaluated by mentors, and turn finished work into reusable experience data.
zjunlp/LightMem
ICLR 2026 memory framework for LLMs and AI agents providing long-term storage, retrieval and update, with reproduction scripts for LoCoMo and LongMemEval and an MCP server.
adolfousier/opencrabs
OpenCrabs is an autonomous self-improving AI agent shipped as a single Rust binary with TUI, CLI and daemon modes, keeping sessions, memory, embeddings and keys entirely on the local machine.
A-EVO-Lab/a-evolve
A-Evolve is a Python infrastructure for self-improving agents: it takes a base agent and a benchmark and runs automated evolution cycles that rewrite the agent without human harness engineering.
framerslab/agentos
AgentOS is an Apache-licensed TypeScript agent framework built around persistent cognitive memory, runtime tool forging in a sandbox, optional HEXACO personality traits and multi-agent orchestration across eleven LLM providers.
YunjueTech/Yunjue-Agent
Self-evolving agent system that starts from zero tools and builds its own during a task, released with a technical report, full system traces and a reproduction branch for five open-ended benchmarks.
allenai/lumos
Lumos is an open-source project that develops modular language agents with unified data formats and competitive performance, leveraging LLAMA-2 and GPT-4 annotations for complex interactive tasks.
baidu-baige/LoongFlow
LoongFlow is a Baidu agent development framework built on a planning-execution-summary loop and a fused memory store, so agents accumulate experience across iterations; it ships general, mathematics and machine learning agents.
alexisfox7/PRO-LONG
Research harness giving long-horizon LLM agents programmatic memory: every observation and action is appended to one structured log that the agent searches with grep and Python, reported at 97.4% on ARC-AGI-3.
shobrook/saplings
Saplings is a library that enables building AI agents with advanced reasoning capabilities using tree search algorithms like Monte Carlo Tree Search, A*, and greedy best-first search for improved decision-making and task performance.
allenai/clin
CLIN is a continually learning language agent designed for rapid task adaptation and generalization within the ScienceWorld environment using advanced language models like GPT-4.
USC-GVL/Agent-Driver
Agent-Driver is an autonomous driving system that uses large language models as cognitive agents to integrate human-like reasoning and common sense into driving, significantly outperforming traditional methods on the nuScenes benchmark.