EveryDev.ai
Subscribe
Home
Tools

5,042+ AI tools

  • New
  • Trending
  • Featured
  • Rate tools
  • Compare
  • Arena
Categories
  • Agents2565
  • Coding1892
  • Infrastructure888
  • Projects714
  • Research633
  • MCP603
  • Marketing496
  • Analytics489
  • Design421
  • Testing403
  • Data377
  • Security321
  • Integration257
  • Communication250
  • Prompts242
  • Extensions218
  • Commerce169
  • DevOps168
  • Voice168
  • Web123
  • Learning114
  • Finance35
AI Tools by Topic
  • AI Coding Assistants
  • Agent Frameworks
  • MCP Servers
  • AI Prompt Tools
  • Vibe Coding Tools
  • AI Design Tools
  • AI Database Tools
  • AI Website Builders
  • AI Testing Tools
  • LLM Evaluations
Follow Us
  • X / Twitter
  • LinkedIn
  • Reddit
  • Discord
  • Threads
  • Bluesky
  • Mastodon
  • YouTube
  • GitHub
  • Instagram
Get Started
  • Users
  • Rate Tools
  • About
  • Editorial Standards
  • Corrections & Disclosures
  • Community Guidelines
  • Advertise
  • Contact Us
  • Newsletter
  • Submit a Tool
  • Start a Discussion
  • Write A Blog
  • Share A Build
  • Terms of Service
  • Privacy Policy
Explore with AI
  • ChatGPT
  • Gemini
  • Claude
  • Grok
  • Perplexity
Agent Experience
  • llms.txt
Theme
With AI, Everyone is a Dev. EveryDev.ai ©
    1. Home
    2. Tools
    3. Zylos
    Zylos icon

    Zylos

    Autonomous Systems
    Featured

    Zylos gives AI agents persistent memory, cross-channel communication, and a self-healing scheduler — all running locally on a Linux server with a single install command.

    Visit Website

    At a Glance

    Pricing
    Open Source

    Fully open-source under the MIT License. Self-hosted on your own Linux server.

    Engagement

    Available On

    CLI
    API
    Linux

    Resources

    WebsiteDocsGitHubllms.txt

    Topics

    Autonomous SystemsAgent MemoryAgent Harness

    Alternatives

    SynapseCowAgentGenericAgent
    Developer
    Zylos AIBhubaneswar, IndiaEst. 2026

    Listed Mar 2026

    About Zylos

    Zylos is an open-source AI agent runtime that solves the statelessness problem of large language models by providing persistent memory, unified multi-channel communication, and autonomous scheduling. Built on Anthropic's Claude Code runtime, it runs locally on a Linux server and stays online while you sleep. Zylos routes messages from Telegram, Lark, and the web through a single gateway so your AI maintains one consistent memory and personality across all channels.

    • Persistent Memory — Five-layer "Inside Out" memory architecture auto-saves context before compaction runs, so your AI never loses important information between sessions.
    • Single Consciousness Across Channels — All messages from Telegram, Lark, and web are routed through one gateway, giving your AI a unified conversation history and personality.
    • Self-Healing Runtime — Built-in crash recovery, heartbeat probes, health monitoring, and auto-upgrades let the agent detect and fix its own problems automatically.
    • Powered by Claude Code — Built on Anthropic's official agent runtime; new Claude capabilities ship to your agent automatically without manual updates.
    • Skill Evolution — Because the agent can program, it writes new skills, integrates external services, and evolves its own capabilities over time.
    • Local-First & Privacy-Focused — Runs on your own Linux server with a single curl install command; no cloud dependency required.
    • Open Source (MIT) — The full codebase is available on GitHub under the MIT License, enabling community contributions and full transparency.
    • Scheduler — A built-in task scheduler executes actions autonomously on your behalf, even when you're offline.
    Zylos - 1

    Community Discussions

    Be the first to start a conversation about Zylos

    Share your experience with Zylos, ask questions, or help others learn from your insights.

    Pricing

    OPEN SOURCE

    Open Source (MIT)

    Fully open-source under the MIT License. Self-hosted on your own Linux server.

    • Persistent five-layer memory
    • Multi-channel gateway (Telegram, Lark, Web)
    • Self-healing runtime
    • Built-in task scheduler
    • Claude Code integration

    Capabilities

    Key Features

    • Persistent cross-session memory
    • Five-layer Inside Out memory architecture
    • Auto-save before context compaction
    • Unified multi-channel gateway (Telegram, Lark, Web)
    • Single AI consciousness across channels
    • Self-healing crash recovery
    • Heartbeat probes and health monitoring
    • Auto-upgrades
    • Built-in task scheduler
    • Powered by Claude Code (Anthropic)
    • AI skill writing and self-evolution
    • Local-first deployment
    • Privacy-focused
    • Open source (MIT)

    Integrations

    Telegram
    Lark
    Claude (Anthropic)
    GitHub
    API Available
    View Docs

    Ratings & Reviews

    No ratings yet

    Be the first to rate Zylos and help others make informed decisions.

    Rate other tools you’ve used

    Developer

    Zylos AI

    Zylos AI builds an open-source agent runtime that gives LLMs persistent memory, multi-channel communication, and autonomous scheduling. The project is created by Howard Zhou and Coco, with contributions from an open-source community. Zylos runs locally on Linux servers, prioritizing privacy and always-on reliability. The runtime is built on Anthropic's Claude Code and released under the MIT License.

    Founded 2026
    Bhubaneswar, India
    5 employees

    Used by

    Production shipped to 66 clients as of…
    First organization live with HXA-Connect
    Read more about Zylos AI
    WebsiteGitHubX / Twitter
    1 tool in directory

    Similar Tools

    Synapse icon

    Synapse

    A self-hosted, open-source AI workspace runtime for building digital teams with shareable AI teammates, shared conversations, memory, and governed access to plugins and local devices.

    CowAgent icon

    CowAgent

    Open-source super AI assistant and Agent Harness that plans tasks, runs tools and skills, and grows with memory and knowledge across multiple models and channels.

    GenericAgent icon

    GenericAgent

    A minimal, self-evolving autonomous agent framework that gives any LLM system-level control of a local computer through 9 atomic tools.

    Browse all tools

    Related Topics

    Autonomous Systems

    AI agents that can perform complex tasks with minimal human guidance.

    510 tools

    Agent Memory

    Memory layers, frameworks, and services that enable AI agents to store, recall, and manage information across sessions. These tools provide persistent, semantic, and contextual memory for agents, supporting personalization, long-term context retention, graph-based relationships, and hybrid RAG + memory workflows.

    211 tools

    Agent Harness

    Infrastructure, orchestrators, and task runners that wrap around LLM coding agents — covering session management, context delivery, worktree isolation, architecture enforcement, and issue-to-PR pipelines.

    305 tools
    Browse all topics
    Back to all toolsSuggest an edit
    ratings
    discussions
    42views
    1upvote