AGENTCONN
H

// Agent profile

holaOS

Coding Free

About holaOS

holaOS is an open-source agent workspace that lets you run any AI agent — Claude Code, Codex, or its built-in agent — across your tools, apps, browser, and files with shared persistent memory. It provides 100+ integrations via MCP, built-in frontier models, and BYOK support. Context, preferences, and project history live in a single shared memory stored as plain local files, so you can switch agents, close the app, and pick up exactly where you left off.

Key Features

  • Run Claude Code, Codex, and built-in agents with shared memory
  • 100+ tool integrations via MCP protocol
  • Persistent local memory stored as plain readable files
  • Built-in frontier models plus BYOK for any provider
  • Browser, file system, and app automation in one workspace
  • Long-horizon agent continuity across sessions

Overview

holaOS is an open-source agent workspace built for long-horizon work, continuity, and cross-agent collaboration. Rather than treating each agent session as isolated, holaOS provides a structured operating environment where multiple agents share the same memory, tools, skills, and application integrations. Context and project history persist locally as plain files you can read and edit, so switching between agents or closing the app does not lose state.

Key Capabilities

The core differentiator of holaOS is its shared persistent memory layer. Claude Code, Codex, and the built-in holaOS agent all operate on the same memory, tools, and application context. The workspace provides 100+ integrations via the MCP protocol, covering browser automation, file management, and third-party app connectivity. Built-in frontier models are available out of the box, and BYOK support lets you connect your own API keys for OpenAI, Anthropic, or any compatible endpoint.

Use Cases

Developers use holaOS for complex, multi-session projects where continuity matters. A typical workflow involves starting a task with one agent, switching to another for a different capability, and having both agents share the same understanding of project state and history. It is also useful for teams standardizing agent tooling across projects, since the shared memory and integrations layer provides consistency regardless of which underlying agent is doing the work.

Considerations

holaOS is free and open-source. Built-in model access may have usage limits, while BYOK mode uses your own API credits. The shared memory model stores everything locally as plain files, which is transparent and portable but means memory management scales with project complexity.

Who It’s For

holaOS is built for developers who want a unified workspace across multiple AI agents with persistent memory and broad tool integration. It appeals to users working on long-running projects who need agent continuity, and to teams who want a single environment that supports multiple agent backends.

Similar Agents