Files
alex e7819bfc82 feat: implement docker infrastructure (US-001)
- Add docker-compose.yml with postgres, redis, api, and web services
- Add multi-stage Dockerfile for API (Python 3.11)
- Add multi-stage Dockerfile for web (Node.js 20 + nginx)
- Add Makefile with common development commands
- Add .env.example with all required environment variables
- Add placeholder pyproject.toml and package.json for builds
- Configure health checks for all services
- Setup persistent volumes for postgres, redis, and repos
- Run services as non-root users
2026-05-16 17:44:39 +00:00

91 lines
2.3 KiB
Markdown

# Tool Instance Management Specification
## Purpose
Launch, monitor, and manage development tool instances in Docker containers.
## Requirements
### Requirement: Tool Instance Creation
The system SHALL create and launch tool instances from repositories.
#### Scenario: Launch tool
- GIVEN an authenticated user with a project and repository
- WHEN they create a tool instance
- THEN:
1. A unique subdomain is generated: `{tool-name}-{tool-id}.hq.local`
2. The Docker Compose template is rendered with project values
3. `docker compose up -d` is executed
4. Container ID and status are stored
### Requirement: Tool Lifecycle
The system SHALL manage tool lifecycle operations.
#### Scenario: Stop tool
- GIVEN a running tool instance
- WHEN the user stops it
- THEN `docker compose stop` is executed
- AND status is updated to "stopped"
#### Scenario: Start tool
- GIVEN a stopped tool instance
- WHEN the user starts it
- THEN `docker compose start` is executed
- AND status is updated to "running"
#### Scenario: Delete tool
- GIVEN a tool instance
- WHEN the user deletes it
- THEN the container and volumes are removed
- AND the database record is deleted
### Requirement: Traefik Integration
The system SHALL auto-generate Traefik labels for routing.
#### Scenario: Route generation
- GIVEN a running tool instance
- THEN these labels are set:
- `traefik.enable=true`
- `traefik.http.routers.{tool_id}.rule=Host(\`{subdomain}.hq.local\`)`
- `traefik.http.routers.{tool_id}.entrypoints=web`
- `traefik.http.services.{tool_id}.loadbalancer.server.port={port}`
### Requirement: Status Monitoring
The system SHALL track tool status.
#### Scenario: Status check
- GIVEN a tool instance
- WHEN status is queried
- THEN the real-time container status is returned:
- pending, building, running, stopped, error
### Requirement: Log Access
The system SHALL provide access to container logs.
#### Scenario: View logs
- GIVEN a tool instance
- WHEN logs are requested
- THEN the last 100 lines are returned
- AND live streaming is available via WebSocket
## Dependencies
- tool-types (tool definitions)
- git-repo (repository access)
- project-management (project context)
- Docker runtime
- Traefik reverse proxy
## Quality Gates
- `pytest` must pass
- `mypy .` must pass
- `ruff check .` must pass
- `npm run typecheck` must pass
- `npm run lint` must pass