OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Workflow Agents

AgentDesk

AgentDesk is an open-source, self-hostable AI customer-support helpdesk that combines web chat, knowledge-base answers, human handoff and ticket workflows with an agent workspace and model configuration.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

AgentDesk — OpenViST project profile in Workflow Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use AgentDesk when you want website support conversations, knowledge-based AI responses, human agent operations and ticket tracking in one self-hosted system.

Who it is for

  • Support teams handling website or SaaS customer questions
  • Organizations that want AI answers grounded in support knowledge with staff escalation
  • Teams managing service requests through conversations, ticket assignment and status tracking
  • Developers adapting a self-hosted helpdesk or embedding its customer-support experience

Documented uses

  • Answer common customer questions using FAQs, documents and other retrievable knowledge.
  • Escalate a conversation to a human when a customer asks, knowledge is insufficient or a workflow needs confirmation.
  • Manage conversations, assign or transfer them, and link customers to support records.
  • Create, categorize, assign and track tickets through status changes and progress records.
  • Configure models and knowledge sources for scenario-specific support agents.
Before you choose

Requirements & limits

The project is open-source under the Apache-2.0 license. Its documented Docker Compose setup provides a default administrator account; change the default password before exposing the system to the public internet or a team environment.

  • The Answerability Gate is intended to route unsupported knowledge-based questions toward fallback and human support.
  • The documented model configuration supports OpenAI-compatible providers and separate language, embedding and reranking model settings.
  • The system supports SQLite or MySQL and uses Qdrant for vector search.
  • Docker Compose starts the application with MySQL and Qdrant; local development additionally requires Go 1.26+, Node.js 20+, pnpm and Qdrant.
  • The repository documents a web widget integration path for adding customer support to a website.

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

Prepare AgentDesk for local development

Prepare the configuration and frontend dependencies, then start the backend and frontend development servers.

Before you start

  • Go 1.26 or later
  • Node.js 20 or later
  • pnpm
  • Qdrant

Steps

  1. Copy the example configuration with cp config/config.example.yaml config/config.yaml.

  2. If Qdrant is not running locally, start it with the command below.

    Documented command
    docker run -p 6333:6333 -p 6334:6334 qdrant/qdrant

  3. Install frontend dependencies by running cd web, pnpm install, and cd ...

  4. Start both development servers with task dev, then use the listed development URLs, such as http://localhost:3000/dashboard or http://localhost:3000/support/chat.

Expected outcome

The project’s local development servers are started, with dashboard, agent workspace, and customer chat URLs listed in the README.

Guide limitations

  • The README lists Qdrant as a requirement and gives a Docker command only for the case where it is not already running locally.
  • This guide covers local development setup, not production deployment or model-provider configuration.
Documentation 1 ↗
Evidence and transparency

Sources & review

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