ИИ-агенты
magic-context
Self-managing, unbounded context and long-term memory for coding agents: one session, for life.
★ 2,3 тыс. звёздMIT
Смотреть руководство →claude-mem — проект с открытым исходным кодом из категории «ИИ-агенты» от thedotmack (thedotmack/claude-mem на GitHub). У него 97 934 звезды на GitHub и 8 612 форков, он распространяется по лицензии Apache-2.0 и последний раз обновлялся на GitHub 7 октября 2026.
Give your coding agents a persistent memory that survives every session.
Be clear about the problem you are solving before installing.
Clone the repository and skim the docs for hardware or key requirements.
Install dependencies, then add API keys, endpoints, or local-model settings.
Launch the example flow and verify the output matches expectations.
Captures session activity and re-injects relevant context later.
Summarizes sessions so memory stays compact and relevant.
Works with Claude Code, Codex, Gemini, Copilot, OpenCode, and more.
Uses ChromaDB as its vector memory store.
A modern TS implementation that is easy to extend.
Permissive open-source licensing.
Download the source and move into the project folder.
git clone https://github.com/thedotmack/claude-mem.git
cd claude-memInstall the TypeScript dependencies.
npm installBuild the project and follow the README to hook it into your agent CLI.
npm run build
# see README for per-agent setupIt is maintained by thedotmack at thedotmack/claude-mem, with 8,612 forks. Review the contributors and release cadence to judge maintenance health.
Yes. It is released under the Apache-2.0 license, which permits commercial use, modification, and redistribution. Read the LICENSE file for the exact terms.
You need model access for the compression and injection steps — typically your existing agent CLI subscription or API keys. Check the README for what it calls.
It is self-hosted and agent-CLI oriented, with local ChromaDB storage. Compare on retention quality, latency, and which agents each option supports.
Clone the repo, run npm install, and follow the README to connect it to the agent CLI you use.
Self-managing, unbounded context and long-term memory for coding agents: one session, for life.
Agentic memory for LLM agents, from a published research project.
Drop-in persistent memory infrastructure for AI agents and apps.
Open-source AI memory that gives agents persistent long-term recall.
Real-time knowledge graphs that give agents a memory with structure.
Self-evolving memory OS for LLMs and agents: persistent graph memory, hybrid retrieval, skill reuse.
Другие проекты из категории «ИИ-агенты» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если claude-mem вам не подходит.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Соседние проекты из категории «ИИ-агенты» в каталоге, ближайшие по числу звёзд к claude-mem — продолжайте изучать полку по обе стороны от него.
Multi-agent LLM framework that mirrors a trading firm: analysts, researchers, trader and risk team
Give your AI agent eyes on the whole internet: Twitter, Reddit, YouTube, GitHub and more, one CLI
A TypeScript toolkit for building agents: unified LLM API, loop, TUI, and CLI.
The official collection of reference Model Context Protocol servers.
The open-source library that lets AI agents control a real browser.
Open-source AI-driven development platform: autonomous software agents in a sandbox
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.