vlmrun-cli-skill

Use the VLM Run CLI (`vlmrun`) to interact with Orion visual AI agent. Process images, videos, and documents with natural language. Triggers: image understanding/generation, object detection, OCR, video summarization, document extraction, image generation, visual AI chat, 'generate an image/video', 'analyze this image/video', 'extract text from', 'summarize this video', 'process this PDF'.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "vlmrun-cli-skill" with this command: npx skills add spillai/vlmrun-cli-skill

VLM Run CLI

Chat with VLM Run's Orion visual AI agent via CLI.

Setup

uv venv && source .venv/bin/activate
uv pip install "vlmrun[cli]"

Environment Variables

You must load the following variables in your environment so that the CLI can use them. You may load the ./env file to your environment.

VariableTypeDescription
VLMRUN_API_KEYRequiredYour VLM Run API key (required)
VLMRUN_BASE_URLOptionalBase URL (default: https://agent.vlm.run/v1)
VLMRUN_CACHE_DIROptionalCache directory (default: ~/.vlmrun/cache/artifacts/)

Command

vlmrun chat "<prompt>" -i input.jpg [options]

Options

FlagDescription
-p, --promptPrompt text, file path, or stdin
-i, --inputInput file(s) - images, videos, docs (repeatable)
-o, --outputArtifact directory (default: ~/.vlmrun/cache/artifacts/)
-m, --modelvlmrun-orion-1:fast, vlmrun-orion-1:auto (default), vlmrun-orion-1:pro
-s, --sessionOptional session ID to continue a previous session
-j, --jsonRaw JSON output
-ns, --no-streamDisable streaming
-nd, --no-downloadSkip artifact download

Examples

Images

vlmrun chat "Describe what you see in this image in detail" -i photo.jpg
vlmrun chat "Detect and list all objects visible in this scene" -i scene.jpg
vlmrun chat "Extract all text and numbers from this document image" -i document.png
vlmrun chat "Compare these two images and describe the differences" -i before.jpg -i after.jpg

Image Generation

vlmrun chat "Generate a photorealistic image of a cozy cabin in a snowy forest at sunset" -o ./generated
vlmrun chat "Remove the background from this product image and make it transparent" -i product.jpg -o ./output

Video

vlmrun chat "Summarize the key points discussed in this meeting video" -i meeting.mp4
vlmrun chat "Find the top 3 highlight moments and create short clips from them" -i sports.mp4
vlmrun chat "Transcribe this lecture with timestamps for each section" -i lecture.mp4 --json

Video Generation

vlmrun chat "Generate a 5-second video of ocean waves crashing on a rocky beach at golden hour" -o ./videos
vlmrun chat "Create a smooth slow-motion video from this image" -i ocean.jpg -o ./output

Documents

vlmrun chat "Extract the vendor name, line items, and total amount" -i invoice.pdf --json
vlmrun chat "Summarize the key terms and obligations in this contract" -i contract.pdf

Prompt Sources

# Direct prompt
vlmrun chat "What objects and people are visible in this image?" -i photo.jpg

# Prompt from file
vlmrun chat -p long_prompt.txt -i photo.jpg

# Prompt from stdin
echo "Describe this image in detail" | vlmrun chat - -i photo.jpg

Continuing a previous session

If you want to keep the past conversation and generated artifacts in context, you can use the -s flag to continue a previous session using the session ID generated when you started the session.

# Start a new session of an image generation task where a new character is generated
vlmrun chat "Create an iconic scene of a ninja in a forest, practicing his skills with a katana?" -i photo.jpg

# Use the previous chat session in context to retain the same character and scene context (where the session ID is <session_id>)
vlmrun chat "Create a new scene with the same character meditating under a tree" -i photo.jpg -s <session_id>

Skipping artifact download

If you want to skip the artifact download, you can use the -nd flag.

vlmrun chat "What objects and people are visible in this image?" -i photo.jpg -nd

Notes

  • Use -o ./<directory> to save generated artifacts (images, videos) relative to your current working directory
  • Without -o, artifacts save to ~/.vlmrun/cache/artifacts/<session_id>/
  • Multiple input files upload concurrently

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

Content Collector

个人内容收藏与知识管理系统。收藏、整理、检索、二创。 Use when: (1) 用户分享链接/文字/截图并要求保存或收藏, (2) 用户说"收藏这个"/"存一下"/"记录下来"/"save this"/"bookmark"/"clip this", (3) 用户要求按关键词/标签搜索之前收藏的内容, (4) 用...

Registry SourceRecently Updated
Coding

Github Stars Tracker

GitHub 仓库 Stars 变化监控与通知。追踪指定仓库的 star 增长、fork 变化,发现新趋势。适合开发者关注项目动态。

Registry SourceRecently Updated
Coding

RabbitMQ client guide for Tencent Cloud TDMQ

RabbitMQ 客户端代码指南。当用户需要编写、调试或审查 RabbitMQ 应用代码时使用。涵盖:用任意语言(Java/Go/Python/PHP/.NET)写生产者或消费者;排查连接暴增、消息丢失、Broken pipe、消费慢、漏消费等客户端问题;审查 spring-boot-starter-amqp、a...

Registry SourceRecently Updated
vlmrun-cli-skill | V50.AI