Vibe-Skills modal-labs

Modal Labs (modal.com) — run Python on serverless containers with GPUs, batch jobs, and autoscaling. Precision wrapper to avoid confusion with UI “modal dialogs”.

install

source · Clone the upstream repo

git clone https://github.com/foryourhealth111-pixel/Vibe-Skills

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/foryourhealth111-pixel/Vibe-Skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/bundled/skills/modal-labs" ~/.claude/skills/foryourhealth111-pixel-vibe-skills-modal-labs && rm -rf "$T"

manifest: bundled/skills/modal-labs/SKILL.md

Modal Labs (modal.com)

Overview

This is a precision wrapper for the upstream

modal

skill (Modal Labs, modal.com). It exists because “modal” is also a common term for UI dialogs (React/Vue/AntD/etc.).

Use this skill only when the user clearly means Modal Labs (modal.com).

When to Use

Route here when prompts mention one or more of:

```
modal.com
```
/ “Modal Labs”
CLI verbs:
```
modal run
```
,
```
modal deploy
```
,
```
modal serve
```
serverless containers for Python, batch jobs, autoscaling
GPU workloads (inference, training, rendering) in a serverless setup

Do not use this skill for UI “modal dialog” tasks.

Setup (CLI)

# Install
uv uv pip install modal

# Login (writes token to ~/.modal.toml)
modal token new

Minimal Example

import modal

app = modal.App("hello-modal")

@app.function()
def hello():
    return "hello from Modal"

@app.local_entrypoint()
def main():
    print(hello.remote())

Run:

modal run script.py

Next Actions (choose based on intent)

One-off run:
```
modal run
```
Long-running endpoint:
```
modal deploy
```
/
```
modal serve
```
GPU function: add
```
@app.function(gpu="H100")
```
(or another GPU type)

If you need deeper patterns (images, volumes, secrets, web endpoints), follow the upstream

modal

skill guidance.