# inference-sh

A bundle of skills that gives your agent image and video generation, LLM calls and web search through the inference.sh CLI.

- plugin · `type=plugin` · Claude Code · ★ 777
- Author: inference-sh
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/inference-sh/skills
- Updated: 2026-10-15
- HTML: https://affpapa.org/en/nevibecode/plugins/inference-sh

## What it does

The plugin drops a set of skills into Claude, invoked as namespaced slash commands: `/inference-sh:flux-image`, `/inference-sh:google-veo` and others. It covers image generation with models like FLUX, Gemini and Reve, video generation with Veo, Seedance and Wan, LLM calls (Claude, Gemini, Kimi, GLM), web search via Tavily and Exa, and X/Twitter automation. Separate groups hold JavaScript and Python SDKs, agent and chat UI components, plus guides on prompting, design, video, writing and social.

You need the inference.sh CLI, called `belt`, and a login through `belt login`. Browse available apps with `belt app list`. Skills will not run without an authenticated CLI, so check that right after installing the plugin.

You do not have to take the whole set. Individual skills install with `npx skills add inference-sh/skills@flux-image`, or `@google-veo`, `@llm-models`, `@web-search`. There is a manual route too: copy the `tools`, `ui`, `sdk` and `guides` directories into `~/.claude/skills/`.

## Install

### Claude Code (native platform)

**Plugin marketplace**
In the agent chat (not a shell):
```text
/plugin marketplace add anthropics/claude-plugins-community
```
In the agent chat (not a shell):
```text
/plugin install inference-sh@claude-community
```

## When to use it
- generate an image from a text prompt with FLUX or Gemini
- turn a prompt into a short video with Veo
- call an LLM from the agent without wiring a separate API client
- pull fresh web results through Tavily or Exa
- build a React chat interface for an agent

## Example prompts
- generate a cover image for this article with /inference-sh:flux-image
- make an 8-second video from this description with /inference-sh:google-veo
- search the web for the latest material on this topic

## Similar
- [extension-inference](https://affpapa.org/en/nevibecode/skills/extension-inference.md) — skill · 11900 installs
- [runcomfy-cli](https://affpapa.org/en/nevibecode/skills/runcomfy-cli.md) — skill · 380100 installs
- [gke-inference](https://affpapa.org/en/nevibecode/skills/gke-inference.md) — skill · 6791 installs
- [mmx-cli](https://affpapa.org/en/nevibecode/skills/mmx-cli.md) — skill · 32000 installs
- [bailian-gen](https://affpapa.org/en/nevibecode/skills/bailian-gen.md) — skill · 9200 installs
- [fal-prompting](https://affpapa.org/en/nevibecode/skills/fal-prompting.md) — skill · 510 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md