Lightning AI Agent Skills

Agent skills that teach AI coding agents (Claude Code, Codex, Cursor, and any agent that supports the SKILL.md format) how to use the Lightning AI platform: GPU Studios, batch jobs, pipelines, LitData dataset workflows, model deployments, code-execution sandboxes, the LLM gateway, durable shareable artifact links, up-front cost estimates for any of it, and billing: balances and spend reports.

Skills

Skill	What it covers
lightning-studios	Create, start, stop and manage cloud GPU Studios; switch machines, run commands, transfer files, SSH
lightning-jobs	Launch and monitor batch jobs (single and multi-machine) on CPUs/GPUs, stream logs, show a live progress bar with ETA and setback tracking, SSH into running jobs or multi-machine workers, collect artifacts
lightning-deployments	Deploy containers/APIs with autoscaling, manage releases, endpoints and auth
lightning-sandboxes	Fast ephemeral VMs for safe code execution: run commands, background processes, file I/O, Docker (docker / docker compose) and public port URLs
lightning-llm-gateway	Call hosted LLMs (OpenAI, Anthropic, open models) through Lightning's models API
lightning-artifacts	Publish a file and get a durable, public lightning.ai/artifacts/<id> link that never expires and renders inline; list, revoke, and delete shares — entirely via the CLI with regular auth
lightning-cost-estimation	Quote what a training run, fine-tune or deployment costs: live per-hour GPU/CPU prices for every cloud, spot rates, multi-node fan-out, and Drive storage
lightning-pipelines	Automate dependent jobs and recurring training; update schedules, diagnose exact runs and clean up resources
lightning-litdata	Turn source data into verified training batches; stream, resume and tune datasets using the curated LitData guides
lightning-billing	Check credit balances and see what was actually spent: by resource type, user, teamspace, day or single job; LLM token spend, monthly credits, plan limits and price quotes; transfer credits and get Stripe top-up links

Tutorial: teach a 4B model to call tools on a cloud GPU. It starts from a Mac or a Claude cloud session and one chat message.

Platform skills use the lightning-sdk Python package and its lightning CLI, plus the raw lightning api escape hatch for anything the SDK doesn't wrap. The LitData skill uses litdata and PyTorch; local dataset processing needs no Lightning account.

Install

Claude Code plugin

This repo is also a plugin marketplace, so Claude Code can install all nine skills as one plugin and keep them updated. The plugin also adds live job progress bars above the prompt. From a shell (no Claude session needed, so it works in scripts and headless setups):

claude plugin marketplace add Lightning-AI/skills
claude plugin install lightning@lightning-ai

Or from inside a Claude Code session:

/plugin marketplace add Lightning-AI/skills
/plugin install lightning@lightning-ai

Plugin skills are namespaced, so they appear as /lightning:lightning-jobs, /lightning:lightning-studios, and so on. They still fire on their own when a task calls for them.

skills.sh (Claude Code, Cursor, Codex and others)

The skills.sh CLI installs into Claude Code, Cursor, Codex, and many other agents:

# interactive: pick skills and target agents
npx skills add Lightning-AI/skills

# install every skill for one agent, user-level, without prompts (swap codex for your agent)
npx skills add Lightning-AI/skills --agent codex --skill '*' --global --yes

# install a specific skill, e.g. just sandboxes
npx skills add Lightning-AI/skills -s lightning-sandboxes

# user-level (global) instead of the current project
npx skills add Lightning-AI/skills -g

Avoid --all unless you mean it: it selects every skill and every agent on the machine.

Manual copy

Or copy the skill folders straight into your agent's skills directory:

SKILLS="lightning-studios lightning-jobs lightning-deployments lightning-sandboxes lightning-llm-gateway lightning-artifacts lightning-cost-estimation lightning-pipelines lightning-litdata lightning-billing"

# Claude Code (project-level)
mkdir -p .claude/skills && cp -r $SKILLS .claude/skills/

# Claude Code (user-level)
cp -r $SKILLS ~/.claude/skills/

Prerequisites

• Python 3.11+ with uv or pip. There's no need to install the SDK first: skills use the lightning CLI in your current environment, installing or upgrading lightning-sdk there if it's missing or too old (uv tool install lightning-sdk or uvx are fallbacks)
• A Lightning AI account for cloud operations (not required for local LitData processing). Sign in once, where the agent runs:
  • On your machine: lightning login (browser sign-in, saved for later commands)
  • Headless (CI, cloud agents, no browser): set LIGHTNING_API_KEY (plus LIGHTNING_USER_ID, optional) through that environment's secrets
  • Inside a Studio: nothing to do, you're already signed in

Important

These skills create real, billed resources on your Lightning AI account (Studios, jobs, deployments, sandboxes). They ask before anything that spends money or makes something public, but you are responsible for what runs under your account.

Conventions

• Skills never guess the organization or teamspace: when more than one is available and none is configured, they ask the user which one to use.
• Skills prefer the documented SDK/CLI surface and fall back to lightning api <endpoint> for raw REST calls.

Contributing

Adding or editing a skill? See CONTRIBUTING.md for the SKILL.md format, the house-style section layout, the conventions above, and how to live-test every command against a control plane before opening a PR.

Support

• Bugs and feature requests: open an issue
• Security problems: see SECURITY.md, and don't file them as public issues
• Changes between versions: CHANGELOG.md

License

Apache-2.0. By contributing you agree your contributions are licensed the same way, and to follow the code of conduct.