Skip to content

Local Setup with Docker Compose

Auto-generated by IssAssist Documentation Bootstrap ยท 2026-06-25 Review this page and remove [TODO: verify] markers as you verify each detail.

Local Setup with Docker Compose

Step 1: Clone the IssAssist-Development Repository

git clone https://github.com/IssTech/IssAssist-Development.git

Step 2: Initialize Git Submodules

The code repositories are submodules, so initialize them:

git submodule update --init --recursive

Step 3: Create the .env-global File

Copy from .env-global.example and fill in your secrets. This file is shared across all agents.

cp IssAssist-Agents/.env-global.example IssAssist-Development/IssaAssist-Agents/.env-global

Step 4: Create Per-Agent .env Files

Each agent has its own GitHub token, so create a new .env file for each:

cp IssAssist-Agents/.env_global.example <agent-name>/

Replace <agent-name> with the name of the agent you're creating an environment for (e.g., pm-agent, architect, etc.).

Step 5: Start the Core Services

Start the core services in detached mode:

docker-compose up -d api webgui tsm-agent

Step 6: Start the AI Agents

Start the AI agents in detached mode:

docker-compose --profile agents up -d

This will start all the agents, including pm-agent, architect, developer, security, and tester.

Step 7: Access the Web GUI

Access the web GUI by navigating to http://localhost:3000 or http://localhost:80.

Alternatively, you can access the API backend at http://localhost:8000.

Step 8: Run a One-Shot Test

Run a one-shot test using the pm-agent agent:

docker-compose run --rm pm-agent

This will execute a single-run test without entering an infinite polling loop.

Step 9: View Logs

View the logs for all containers in real-time:

docker-compose logs -f

Step 10: Stop Everything

Stop everything by shutting down the containers and removing them from Docker:

docker-compose down

Note: This guide assumes you're using Docker Compose version 1.x. If you're using a newer version, some commands might be different.