Apache-2.0 multi-agent coordination platform — run persistent, heterogeneous agent systems with shared context, governed execution, and a coordination network of discovery, team formation, routing, and collaboration. Connects via plugin or gateway integration.
DSH integration
Ecosystem-related
Author-claimed
Safety audit
Unaudited
Last verified
2026-09-04
License
Apache-2.0
01What can it help you accomplish?
Run persistent, coordinated, multi-agent systems across applications, runtimes, and human-agent workflows
A local Avernet stack with an Avernet process, a frontend workbench, and 5 local test bots for agent onboarding and coordination
Teams building multi-agent systems who need shared context, governed execution, and long-lived agent collaboration
Connect heterogeneous agent runtimes, plugins, and bot platforms into one collaboration network
A coordination network with discovery, team formation, routing, and collaboration across multiple agents
Platform teams integrating OpenClaw, local agent runtimes, custom bot processes, or existing bot platforms
02How to install into DeepSeek Harness
Installation steps
- 01
Clone the repository: `git clone https://github.com/inclusionAI/Avernet.git` and `cd Avernet`
$ git clone https://github.com/inclusionAI/Avernet.git
- 02
Run `./scripts/singlebox.sh install-tools` to install the required toolchain
- 03
Run `./scripts/singlebox.sh` to start the local Avernet stack
Verify the integration
- Open the frontend workbench at `http://127.0.0.1:8000/`
03DSH integration and capability boundaries
Connects agents into one collaboration network via a plugin/runtime path (agents register, onboard, receive messages, and report results through a plugin) or a gateway path for existing bot platforms.
Agent coordination network
heterogeneous agent runtimes, plugins, and bot platforms→discovery, relationship building, team formation, routing, collaboration, and governance across multiple agents
Trusted core
agents and participants→identity, auth, permissions, security, audit, and lifecycle management
Execution infrastructure
heterogeneous agent engines and bot-as-a-service runtimes→containers, clusters, and operational runtimes for running agent systems
runs local containers, clusters, and operational runtimes on the host
04Who is it for? When not to use it?
Good for
- Teams building multi-agent systems who need shared context, governed execution, and long-lived agent collaboration
- Platform teams integrating OpenClaw, local agent runtimes, custom bot processes, or existing bot platforms
Not for
- Do not commit secrets, tokens, cookies, private keys, private service endpoints, local databases, runtime logs, or machine-specific configuration; if credentials were already committed, revoke or rotate them before cleaning repository history.
05Compatibility, maintenance and safety notes
- Do not commit secrets, tokens, cookies, private keys, private service endpoints, local databases, runtime logs, or machine-specific configuration; if credentials were already committed, revoke or rotate them before cleaning repository history.
- The public demo is not intended to fully demonstrate all production-scale properties of Avernet, such as large-scale connection envelopes, permission isolation, audit depth, failure recovery, or long-horizon organizational collaboration.
Apache-2.0; latest release v2026.08.04 (2026-08-05); pushed 2026-09-03
06Frequently asked questions
How do I install Avernet for DeepSeek Harness?
Clone the repository with `git clone https://github.com/inclusionAI/Avernet.git`, then run `./scripts/singlebox.sh install-tools` followed by `./scripts/singlebox.sh` to start a local stack. Open the workbench at `http://127.0.0.1:8000/`.
How does Avernet integrate with agent engines?
Avernet supports a plugin/runtime path where agents actively connect through a plugin for registration, onboarding, message receiving, and result reporting, and a gateway path that dispatches tasks to an existing bot platform.
What capabilities are available in the public repo?
Trusted core (identity, auth, permissions, lifecycle), agent coordination network (discovery, team formation, routing, collaboration), and execution infrastructure are largely available; security, audit, context, memory, and orchestration are planned or in progress.
Is Avernet open source and what is its license?
Yes — it is licensed under the Apache License 2.0.
What does the public demo show?
The demo shows local onboarding and coordination flow, workbench interaction, and integration of local test bots; it is not intended to fully demonstrate production-scale properties like permission isolation or failure recovery.
07Related DSH workflows
archify
by tt-a1i
Agent skill for beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams—self-contained HTML with motion and crisp export.
openviking
by volcengine
Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
nocobase
by nocobase
NocoBase is an open-source AI + no-code platform for building business systems fast. Instead of generating everything from scratch, AI works on top of production-proven infrastructure and a WYSIWYG no-code interface, so you get both speed and reliability.
learn-harness-engineering
by walkinglabs
Harness engineering beginner tutorial, from 0 to 1
08Data and sources
Avernet does not lock you into a single agent engine. It supports two integration paths for connecting agents, runtimes,…
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-09-04. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.