OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Infrastructure

Hybridclaw

Hybridclaw is a self-hosted AI assistant runtime that combines a gateway, tools, memory and user interfaces with sandboxed execution, credential controls and approval policies.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Hybridclaw — OpenViST project profile in Agent Infrastructure.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Hybridclaw when you need a self-hosted assistant runtime for business workflows and want execution controls, credential handling and memory in the same operational stack.

Who it is for

  • Teams building assistants for CRM, marketing, analytics, finance, cloud, office and operations workflows.
  • Operators who want to host an assistant runtime and manage its credentials, approvals and execution boundaries.
  • Teams coordinating local agents, hosted proxy agents or trusted Hybridclaw peers.

Documented uses

  • Run an assistant that acts on business services through structured skills and tools.
  • Keep secrets outside model context and require approvals for configured actions.
  • Operate local memory and connect agents across installations through encrypted A2A messages.
  • Expose assistant functions through web chat, a terminal interface, an admin console, a scheduler or an OpenAI-compatible API.
Before you choose

Requirements & limits

The repository documents self-hosted installation options and a HybridAI Cloud launch option. Node.js 22 is required for the documented CLI setup; Docker is recommended for the default sandbox. The project is MIT-licensed.

  • Node.js 22 is a stated requirement; Docker is recommended for the default sandbox.
  • Business skills cover CRM, marketing, analytics, finance, cloud, office and operations, with named integrations including Salesforce, HubSpot, Airtable and Microsoft 365.
  • The runtime supports several interfaces and an OpenAI-compatible API, so deployment decisions can account for how users and systems will access it.
  • Memory can use local files, SQLite and semantic recall, with optional HybridAI cloud memory.
  • The README identifies Qwen/Qwen3.6-27B-FP8 as its small-model validation baseline; this does not establish results for other model and workflow combinations.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and start Hybridclaw with npm

Install the CLI, complete onboarding and start the gateway and terminal interface.

Before you start

  • Node.js 22
  • Docker recommended for the default sandbox

Steps

  1. Install the package globally with the command below.

    Documented command
    npm install -g @hybridaione/hybridclaw

  2. Run hybridclaw onboarding.

  3. Start the service with hybridclaw gateway.

  4. Open the terminal interface with hybridclaw tui.

Expected outcome

The CLI is installed, onboarding has been run, and the gateway and TUI start commands have been issued.

Guide limitations

  • The repository recommends Docker for the default sandbox.
  • These documented steps are not an execution test.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
HybridAIOne/hybridclaw · Official homepage title · retained in its source language
How we review sourcesSuggest a correction