lablupOfficial·18 Agent Skills Included

backend.ai

Run and scale multi-tenant AI compute clusters with GPU support

Manages container-based AI computing clusters with scheduling, GPU allocation, and multi-tenant sessions. Exposes every function through REST and GraphQL APIs plus a Python SDK and CLI for automation. Removes the manual work of provisioning kernels, storage volumes, and user permissions across nodes. Includes agent skills for testing, migrations, releases, and local development workflows.

672|181

↗

npx skills add lablup/backend.ai --all -g -y

Available:

Gives the AI agent global coding rules, layer architecture constraints, and a routing index that maps each development task to the correct skill in this repository.

All Skills in This Repository (18)

Pure Emerald Level Indicators

📦 In Repo

lablup

submit

Automates quality checks, commits, PR creation, and changelog generation for Backend.AI contributions.

bep-guide

Guides writing and managing Backend.AI Enhancement Proposals with structured workflows.

release

Automates Backend.AI version releases with towncrier changelog generation and RC consolidation.

test-guide

Guides scenario-first test writing for Backend.AI using pytest fixtures and pants.

halfstack

Diagnose and fix Docker Compose halfstack development environment issues.

observability

Query Backend.AI logs, metrics, and traces through the Grafana MCP during local development.

local-dev

Manages Backend.AI local services via tmux with startup-crash debugging and restart workflows.

knowledge

Search and author KNOWLEDGE.md background-knowledge documents with schema validation scripts.

code-trace

Trace features across REST, GraphQL, Service, Repository, and database layers in Backend.AI.

bai-cli

Tests and verifies Backend.AI REST API endpoints using the v2 ./bai CLI command reference.

cli-sdk-guide

Implement Backend.AI client SDK and CLI for automated API interactions.

tdd-guide

Guide TDD workflows for Backend.AI with pytest and with_tables patterns.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install Backend.AI skills?▼

Run `npx skills add lablup/backend.ai --all -g -y` in your terminal to install all skills in this suite globally.

What is Backend.AI used for?▼

It is an open-source platform that runs multi-tenant AI and machine learning workloads in containers, allocating GPUs and other accelerators across users with its own scheduler.

How do AI agents work with the Backend.AI codebase?▼

The root AGENTS.md file routes each task to a dedicated skill, covering API design, repositories, testing, database migrations, releases, and local service management.

Does Backend.AI support GPUs and other accelerators?▼

Yes. It ships with pluggable accelerator support for NVIDIA CUDA, ROCm, Google TPU, Graphcore IPU, and several NPU vendors.

Can I manage Backend.AI from the command line?▼

Yes. The bundled bai CLI and Python SDK expose every entity operation, from sessions and vfolders to users, deployments, and resource policies.