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

Harnessrouter

Harnessrouter is Apache-2.0 self-hosted infrastructure that exposes multiple agent harnesses through one API, implementing the Unified Harness Protocol and providing persistent sessions, streaming progress, files, artifacts, cancellation and structured failure handling.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Harnessrouter — 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 it when an agent product needs a common API across harnesses or when the routing layer, provider keys and task data should remain on infrastructure controlled by the deployer.

Who it is for

  • Teams building agent products that need one integration point for several harness backends.
  • Developers operating agent infrastructure on their own servers or cloud instances.
  • Builders needing shared session, file, streaming and task-control capabilities across supported harnesses.

Documented uses

  • Run tasks through supported agent harnesses.
  • Retrieve task results and switch between harness backends.
  • Preserve agent sessions and stream progress to a client.
  • Submit files to harness-backed tasks and retrieve artifacts.
  • Add cancellation and structured failure handling to an agent application.
Before you choose

Requirements & limits

The Community Edition is self-hosted and open source under Apache-2.0. The documented local setup requires Docker, about 4 GB of disk space and a provider API key; it states that no Harnessrouter account or bundled model key is required.

  • Supported harnesses include Codex, Claude Code, Hermes, PI and DSH, according to the repository description.
  • The common API covers sessions, streaming, files, artifacts, cancellation and structured failures.
  • The setup installs enabled harness CLIs during first launch, so the available backends depend on the configured installation.
  • The Docker quickstart uses a persistent volume for the database, files, installed CLIs and workspaces.
  • The repository identifies Python as its primary language and describes an OpenAI Responses-compatible API.

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

Start Harnessrouter Community Edition locally

Set up the self-hosted routing layer and open its console.

Before you start

  • Docker
  • About 4 GB of disk space
  • A provider API key

Steps

  1. Run the documented Docker command with the Harnessrouter image, port mapping and persistent volume.

  2. Follow the container logs until they show that Harnessrouter is ready on port 3000.

  3. Open the local console or the selected host port and sign in with the initial credentials.

  4. Change the default password in Profile before using the instance beyond a local setup.

Expected outcome

A local Harnessrouter Community Edition instance is started with persistent data, files, installed harness CLIs and workspaces.

Guide limitations

  • The quickstart does not include a bundled model or trial key.
  • The first launch installs enabled harness CLIs and may require waiting for the console to finish starting.
  • The documented initial credentials should be changed before broader exposure.
Documentation 1 ↗
Evidence and transparency

Sources & review

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