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

Openrath

Openrath is a Python framework and runtime for composing dynamic multi-agent and multi-session workflows through explicit objects for sessions, agents, workflows, tools, memory, selectors and execution sandboxes.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

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

Practical overview

When it fits

Use Openrath when an agent application needs explicit session lineage, reusable workflow components, persistent memory, model-facing tools or runtime routing. Its production profile is relevant when runs must survive worker or process interruptions and operational concerns such as leases, side effects, approvals and tenancy need defined boundaries.

Who it is for

  • Python developers building multi-agent systems.
  • Teams composing self-hosted agent workflows with persistent state.
  • Engineers evaluating durable execution and operational controls for agent runtimes.

Documented uses

  • Compose multiple agents and workflows into larger systems.
  • Carry conversation and inter-agent collaboration state across branches or sessions.
  • Persist agent memory across workflow runs.
  • Expose callable tools to models with defined runtime behavior.
  • Route between workflows dynamically with model-backed selectors.
Before you choose

Requirements & limits

The project is described as open source under the BSD-3-Clause license. The repository presents Python package, CLI and self-hosted usage across Linux, macOS and Windows. The production profile is installed with the server and PostgreSQL extras, followed by a separate schema migration operation.

  • Session is the central flowing runtime value, carrying conversation state and collaboration lineage.
  • Agent, Workflow and Selector provide separate layers for transformation, composition and runtime routing.
  • Memory persists state across runs, while Tool exposes callable operations to models.
  • Sandbox and Backend represent the execution environment for tools, including local-process and OpenSandbox options described by the project.
  • The v2 production path adds immutable plans, durable Runs, Events and Checkpoints, leases, retries, cancellation and resumable queues.

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 the production profile and prepare its schema

Set up Openrath’s documented server and PostgreSQL profile before using its production runtime path.

Before you start

  • Python package installation capability.
  • PostgreSQL support for the production profile.
  • A runtime identity with appropriate operational permissions; runtime identities do not need DDL privileges.

Steps

  1. Install the server and PostgreSQL extras with pip.

  2. Run the Openrath schema migration as a separate operation.

  3. Run the migration check command to inspect migration status.

Expected outcome

The production package is installed and the schema migration operation has been run and checked.

Guide limitations

  • This recipe documents installation and migration only; it does not establish that an agent workflow has been executed.
  • The Agent Server HTTP surface is identified as Beta.
  • The documented production path requires separate operational components including PostgreSQL; Redis and S3-compatible storage are described as supporting services rather than installation steps.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
3 October 2026
Original website title
Rath-Team/OpenRath · Official homepage title · retained in its source language
How we review sourcesSuggest a correction