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

Uniemployee

Uniemployee is an open-source enterprise platform for configuring and running AI digital employees around professional experience, business workflows and decision criteria. It connects employee roles with SOPs, skills, tools, knowledge sources and MCP connectors, and supports human approval stages and execution traces.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Uniemployee — 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 Uniemployee when a team needs to turn documented procedures into configurable AI-assisted roles and wants approval checkpoints and run-level traces in the same platform.

Who it is for

  • Organizations that want to encode repeatable business procedures and decision criteria into AI-assisted roles.
  • Teams configuring AI employees for customer service, sales, HR, data analysis, business analysis or network operations.
  • Technical teams able to run a self-hosted Python application and configure a compatible model endpoint.

Documented uses

  • Configure an AI employee’s role, model, skills, tools, knowledge sources, SOPs and connectors.
  • Represent multi-step business procedures as workflow-driven skills or StateGraph processes.
  • Pause higher-risk workflow actions for human approval, then continue after a decision.
  • Review or replay execution records that capture model and tool activity, timing and token use.
  • Use the analyst workspace to work with database, spreadsheet, knowledge-base and connector data and produce HTML reports.
Before you choose

Requirements & limits

The project documents a self-hosted setup. Its repository identifies the project as MIT licensed.

  • The documented setup requires Python 3.12 or later, PostgreSQL and an OpenAI Chat Completions-compatible model endpoint with an API key.
  • Node.js 18 or later is needed for front-end development mode; the documented single-process service includes the front-end static files.
  • Employees can be assembled from skills, SOPs, tools, knowledge bases and MCP connectors, while StateGraph workflows can include approval nodes.
  • The platform documents web chat and an internal IM channel framework; named external messaging channels remain planned.
  • The analyst workspace and ontology capabilities are specialized parts of the platform, not requirements for every configured employee.

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

Set up Uniemployee locally

Prepare the documented local application environment and open the platform.

Before you start

  • macOS, Linux or Windows
  • Python 3.12 or later
  • Docker for PostgreSQL, or an existing PostgreSQL instance
  • An OpenAI Chat Completions-compatible model endpoint and API key
  • Node.js 18 or later only for front-end development mode

Steps

  1. Clone the repository, enter its directory and copy .env.example to .env.

  2. Set the model name, compatible API base URL, API key and a replacement JWT secret in .env.

  3. Create a Python virtual environment and install dependencies from backend/requirements.lock.txt.

  4. Start PostgreSQL with the command below, then start the service using the documented Uvicorn command on port 8787.

    Documented command
    docker compose up -d db

  5. Open the local service at port 8787 and sign in with the documented initial administrator credentials; change the password when prompted.

Expected outcome

A locally running Uniemployee service with its web interface available at port 8787.

Guide limitations

  • Model hardware requirements depend on the selected model service.
  • The documented default administrator account requires a password change at first login.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
zj-unicom-ai/UniEmployee · Official homepage title · retained in its source language
How we review sourcesSuggest a correction