zylos-ai/zylos-core

🐙 Give your AI a life — open-source agent infrastructure for team collaboration.

What it solves

Zylos provides a persistent, autonomous AI agent system that overcomes the "amnesia" of standard LLM sessions. It transforms a stateless LLM into a long-term companion with a permanent memory, a scheduler for autonomous tasks, and the ability to communicate across multiple platforms (Telegram, Lark, Web Console) while maintaining a single, unified consciousness.

How it works

Zylos acts as a "life system" wrapper around AI runtimes like Claude Code or OpenAI Codex. It uses a C4 Communication Bridge to route all messages from various channels into a single gateway, ensuring the AI maintains one conversation and one personality. It implements an "Inside Out" memory architecture (divided into identity, state, references, sessions, and archive) to prevent data loss during context compaction. To ensure reliability, it includes an Activity Monitor for crash recovery and a Scheduler for autonomous task dispatch.

Who it’s for

Users who want an always-on, autonomous AI agent that can program, evolve its own skills, and be accessed from anywhere without paying per-token API costs (by leveraging existing Claude or Codex subscriptions).

Highlights

  • Unified Consciousness: One AI identity across Telegram, Lark, and web interfaces.
  • Inside Out Memory: A five-layer tiered memory system that prevents memory loss during context window compaction.
  • Self-Healing: Built-in heartbeat probes and crash recovery to keep the agent running autonomously.
  • Cost Efficient: Runs on flat-rate subscriptions rather than per-token API billing.
  • Runtime Flexibility: Supports interchangeable runtimes (Claude Code and Codex) while preserving memory and skills.
  • OpenClaw Compatibility: Integrates with the OpenClaw ecosystem and HXA-Connect B2B protocol for multi-agent collaboration.

Related

  • Project
    CronusL-1141/AI-company

    A shared operating layer for Claude Code and Codex that maintains project memory, tasks, and coordination across AI agent sessions.

  • Project
    danielmiessler/LifeOS

    An AI-powered life operating system that adds persistent memory, personal context, and specialized skills to AI coding harnesses to help users achieve personal and goal-oriented outcomes.

  • Project
    GeminiLight/MindOS

    A local-first knowledge base that provides a shared, durable context for humans and AI agents to collaborate and evolve their shared memory.

  • Project
    sopaco/cortex-mem

    A Rust-based long-term memory framework for AI agents that uses a three-tier hierarchical architecture and a virtual filesystem to provide persistent, context-aware interactions.

  • Project
    mnemon-dev/mnemon

    Mnemon is a cross‑platform CLI that adds persistent, graph‑based memory to LLM agents. The host LLM supervises what to remember, link, or forget, while Mnemon handles deterministic storage, decay, deduplication, and optional vector search. It integrates with many AI coding tools (Claude Code, Codex, Cursor, etc.) via one‑click setup, requiring only a single local binary and no extra API keys.