infsh-cli

Execute over 150 cloud-hosted AI applications via a command-line interface.

688|95|Updated Jan 31, 2026

Search Tags:#automation #serverless #models #cli #ai #inference

Ready to run

npx skills add https://github.com/inference-sh/skills --skill infsh-cli

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: infsh-cli
Source: https://github.com/inference-sh/skills/tree/main/tools/infsh-cli
Command: npx skills add https://github.com/inference-sh/skills --skill infsh-cli

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a command-line interface to access and run a vast array of over 150 AI applications without requiring local GPU resources.

Core Features & Use Cases

Unified Access: Interact with diverse AI models for image generation, video creation, LLM calls, web search, and more through a single CLI.
Simplified Workflow: Execute complex AI tasks with simple commands, abstracting away the complexities of individual model APIs.
Use Case: Generate an image of a "cyberpunk cat" using the FLUX model, then immediately use the output to create a short video clip with the Veo model, all from your terminal.

Quick Start

Run the infsh app run falai/flux-dev-lora --input '{"prompt": "a cat astronaut"}' command to generate an image.

Frequently Asked Questions about infsh-cli

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

FAQPage Schema

How do I run AI models from the command line without a local GPU?▼

You can execute cloud-hosted AI applications via a command-line interface to run tasks like image generation and LLM interactions without local GPU resources. This approach abstracts individual model APIs, enabling serverless task execution directly from your terminal.

Can I generate an image and use it for video creation in one CLI workflow?▼

Yes, you can generate an image using a model like FLUX and immediately use the output to create a video with a model like Veo. This simplified workflow chains multiple cloud-hosted AI tasks sequentially using simple terminal commands.

What is the syntax for executing serverless AI tasks via CLI?▼

Serverless AI tasks are executed using specific app identifiers and JSON input for configuration. You run a command like `infsh app run <app-identifier> --input '<json-data>'` to pass prompts and parameters directly to the cloud-hosted application.

Does command-line AI inference support web search and social media automation?▼

Command-line AI inference supports web search, social media automation, image generation, video creation, and LLM interactions. It provides unified access to over 150 diverse cloud-hosted applications through a single terminal interface.

Do I need to install specific model dependencies to run AI apps locally?▼

You do not need to install model dependencies or have local GPU resources. The command-line interface connects to cloud-hosted applications, abstracting away the complexities of individual model APIs and entirely eliminating local setup requirements.