devops-skill

Runs and deploys an AI agent platform locally or with Docker Compose and PostgreSQL.

Updated Mar 12, 2026
One-click install
npx skills add https://github.com/tendercoconut174/ai-agent-platform --skill devops-skill-tendercoconut174
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: devops-skill
Source: https://github.com/tendercoconut174/ai-agent-platform/tree/main/.cursor/skills/devops-skill
Command: npx skills add https://github.com/tendercoconut174/ai-agent-platform --skill devops-skill-tendercoconut174

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Setting up and running a multi-service AI agent platform involves coordinating a gateway, an orchestrator, and PostgreSQL, which is error-prone without clear operational guidance. This Skill provides the exact commands and configuration needed to run, test, migrate, and deploy the platform locally or with Docker. ## Core Features & Use Cases - Local Development Setup: Start the gateway and orchestrator with uv, either in minimal in-memory mode or full-stack mode backed by PostgreSQL. - Docker Compose Orchestration: Launch all services (gateway on 8000, orchestrator on 8001, postgres on 5432) with a single command and a documented service table. - Database Migrations: Apply, generate, and roll back Alembic migrations through the project CLI. - Use Case: A developer clones the repository, needs to verify the message pipeline end-to-end, and uses this Skill to start PostgreSQL, run migrations, launch both services, and send a test curl request to the gateway. ## Quick Start Ask the agent to start the full stack locally with PostgreSQL and send a test message to the gateway endpoint.

Frequently Asked Questions about devops-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run the AI agent platform locally?▼

Run the orchestrator and gateway in separate terminals using uv run python main.py orchestrator and uv run python main.py gateway. For full functionality, first start PostgreSQL with docker compose up -d postgres and apply migrations with uv run python main.py migrate.

How do I start all services with Docker Compose?▼

Run docker compose up -d to start the gateway on port 8000, the orchestrator on port 8001, and PostgreSQL on port 5432. All services use 8.8.8.8 and 8.8.4.4 for external DNS resolution.

Can I run the project without PostgreSQL?▼

Yes, a minimal mode runs without PostgreSQL by starting only the orchestrator and gateway. Sessions are stored in memory and are lost when the services restart.

How do I run database migrations with Alembic?▼

Apply all migrations with uv run python main.py migrate. Check the current state with uv run alembic current, generate a new migration with uv run alembic revision --autogenerate, and roll back one step with uv run alembic downgrade -1.

What environment variables does the platform require?▼

OPENAI_API_KEY is required for LLM features, with OPENAI_MODEL defaulting to gpt-4o-mini. DATABASE_URL, ORCHESTRATOR_URL, and FILE_WORKSPACE have local defaults and are loaded from a .env file via python-dotenv.

Why is my gateway not connecting to the orchestrator in Docker?▼

In Docker, ORCHESTRATOR_URL must be set to http://orchestrator:8001 instead of the localhost default. Verify both containers are running via docker compose and that the environment variable is set in your .env file.