From docker compose up to Your First Custom Agent

Описание и применение

Запуск платформы Auto Learning Agents в Docker позволяет быстро развернуть локальный AI‑агент с минимальными настройками.

Задача

Run an AI agent platform locally without renting cloud services.

Результат

The platform runs with minimal setup, allowing users to create and schedule custom agents that execute tasks on a defined schedule.

Рабочий процесс

• Clone the repo and copy the example env file
• Run docker compose up -d to start the container
• Verify ports 80 and 9500 are exposed
• Edit agents.json to add a custom agent
• Set tick_minutes, model, and instructions_file
• Enable the agent and observe runs in the activity feed

Доказательства

The platform ships as a single Docker image. It bundles the Elixir/OTP release that supervises every agent, PHP for the tool layer, Python with a local embedding model, Playwright with Chromium for browser automation, and Apache serving the web UI.

Источник

git clone https://github.com/AIAppsAPI/auto-learning-agents cd auto-learning-agents cp .env.example .env docker compose up -d

Источник

A custom agent is one entry in agents.json: ... tick_minutes is the schedule, model is the brain, and the instructions file is the job description.

Источник