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
ghostwright/phantom
Phantom is a TypeScript agent platform whose premise is giving an AI agent its own dedicated computer rather than running it as a disposable chat session. A Phantom instance persists across sessions, keeps long-term memory in a Qdrant vector store with embeddings served by Ollama, installs software and spins up databases on its own machine, creates new tools for itself, and registers them so future sessions and other agents can use them. It is reached through Slack, Telegram, email and webhooks, plus a built-in web chat at the /chat route, and it has its own email address for sending mail through Resend. The README documents concrete behaviour from production instances: one installed ClickHouse on its VM, loaded 28.7 million rows of Hacker News data, built an analytics dashboard and exposed a REST API that it then registered as an MCP tool; another was asked about Discord support, explained the Discord Bot API, guided the user through creating an application, accepted the token through a secure magic-link submission flow, and started the container to permanently gain a channel it did not ship with; a third integrated a small open-source system monitor into its ClickHouse instance and built a self-monitoring dashboard with a thirty-second sync pipeline. Model backends are pluggable through a single YAML block, with support for Anthropic as the default plus Z.AI, OpenRouter, Ollama, vLLM, LiteLLM and any endpoint compatible with the Anthropic Messages API; both the main agent and the self-evolution judges use the configured provider. Deployment is via Docker Compose with an environment file holding the model API key, Slack tokens and owner identifier, and a health endpoint on port 3100. The README warns that the compose file mounts the Docker socket so the agent can spawn sibling containers, which grants root-equivalent access to the daemon, and recommends running it on a dedicated machine.
https://github.com/ghostwright/phantom
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.
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.