skills.rest
JamesFincher
JamesFincherCOMMUNITY
Explore all skills by this author →
GitHub

inference-sh-cli

Discover app IDs and run inference.sh AI applications with JSON output.

Updated May 4, 2026
Search Tags:#authentication#cli#image-generation#video-generation#inference.sh#terminal-tool#llm-search
One-click install
npx skills add https://github.com/JamesFincher/gengar --skill inference-sh-cli-jamesfincher
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: inference-sh-cli
Source: https://github.com/JamesFincher/gengar/tree/main/optional-skills/devops/cli
Command: npx skills add https://github.com/JamesFincher/gengar --skill inference-sh-cli-jamesfincher

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It solves the problem of getting access to many different AI applications without manually managing provider APIs, models, and authentication.

Core Features & Use Cases

  • Unified AI app execution: Runs inference.sh apps (images, video, search, TTS, avatars, and more) through the infsh CLI using the agent’s terminal tool.
  • App discovery first: Finds the correct app ID via search/listing before running, avoiding breakage from renamed or changing IDs.
  • Machine-readable outputs for media: Uses --json output so the agent can surface generated media URLs as MEDIA:<url> for inline display.
  • Local file support: Uploads local images/audio/files by passing local paths directly in JSON inputs when the target app accepts URL fields.

Quick Start

Ask the agent to run inference.sh app search for "flux" and then generate an image prompt like "sunset over mountains" using the best matching FLUX app ID, returning results with MEDIA:<url>.

Frequently Asked Questions about inference-sh-cli

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

FAQPage Schema
How do I run AI image and video generation from the terminal?▼

Run AI image and video generation from the terminal by using the inference.sh CLI to discover hosted app IDs and execute them with structured JSON inputs that output media URLs.

Can I use local files for cloud-based AI generation apps?▼

Yes, you can use local files for cloud-based AI generation by passing local image or audio file paths directly into the app's JSON input fields when the target app accepts URL inputs.

Do I need to manually find app IDs before running inference.sh CLI?▼

You do not need to manually find app IDs because the inference.sh CLI always performs an app lookup or search first to select stable IDs, preventing breakage from renamed or changing applications.

What is the best way to handle authentication for unified AI app execution?▼

The best way to handle authentication for unified AI app execution is to ensure the inference.sh CLI is properly authenticated beforehand, allowing seamless access to multiple hosted provider APIs and models.

How does the inference.sh CLI return generated media for downstream use?▼

The inference.sh CLI returns generated media by using the --json output flag, which surfaces media URLs as MEDIA:<url> for inline display and downstream URL handling.

Why should I use a unified CLI instead of managing individual AI provider APIs?▼

Using a unified CLI avoids manually managing individual provider APIs, models, and authentication, giving you access to many different AI applications for search, TTS, and media generation through a single terminal tool.

Related Skills

Explore More Software Engineering Skills→
MarbleSodasMarbleSodas

inference-sh-cli

Run cloud-based AI apps via the inference.sh CLI with JSON outputs.

Community
Intermediate
feed-mobfeed-mob
3

csv-url-parser

Extracts query parameters from URLs in CSV files into new columns.

Official
Basic
projectedanxprojectedanx
1

inference-sh-cli

Execute over 150 cloud-based AI applications from a unified CLI.

Community
Intermediate
zeronx798zeronx798

shop

Search Shopify's global catalog, manage checkout and track orders via Shop CLI or MCP API.

Community
Advanced