Install a Skill
Send this prompt to your Agent and let it find the Skill you need:
Install this skill https://vtrix.ai/skills/skill.md and follow the instructions to search Skills in vtrix skillhub1
Find a Skill
Browse Skill Hub, find the Skill you want, and click to open its detail page
2
Select your Agent
Vtrix supports Claude Code, Cursor, Windsurf, Cline, Gemini CLI, Codex CLI, OpenCode, GitHub Copilot, Roo Code, Kiro, Junie, Augment Code, Warp, Goose, and more
3
Select your platform
Choose your operating system: macOS, Windows, or Linux
4
Copy the install command
The page provides a one-click install command — click Copy and run it in your terminal
Vtrix Multimodal Skills
Vtrix multimodal Skills typically include the following sections:
# Frontmatter
---
name: image-to-video
## The basis on which the agent decides whether to invoke this Skill — includes a description and trigger keywords
description: animate still images into videos... triggers: image to video, animate image, ...
allowed-tools: Bash(curl *)
metadata:
## List of Vtrix model IDs used by this Skill
vtrix_models:
- kirin_v3_i2v
version: "1.0"
---| Section | Description |
|---|---|
| Model Selection | When a Skill supports multiple models, lists each model’s use case and ID so the agent can choose appropriately |
| Prerequisites | Environment variable setup, typically export VTRIX_API_KEY=your_key_here |
| Quick Start | The minimal runnable example to get started immediately |
| Request Parameters | Full parameter list with types and descriptions |
| Examples | Multiple examples covering different scenarios such as portraits, landscapes, and products |
| Prompt Tips | Prompt writing guidance specific to this Skill |
| Error Handling | Common error codes and how to resolve them |
Once a Skill is loaded, the agent reads its body to understand how to construct API requests. The trigger keywords in description help the agent determine when to load the Skill.