Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Version 0.1.1 of vlmrun-cli-skill - No file changes detected in this release. - Documentation and usage instructions remain unchanged.
What This Skill Does
A CLI tool that sends images, videos, and documents to VLM Run's Orion visual AI agent using natural language prompts. Supports object detection, OCR, image and video generation, and document extraction. Sessions can be continued across multiple interactions to preserve context and generated artifacts.
Processes visual content and generates images or videos from the command line without writing any code, with session continuity that retains generated artifacts and conversation context across follow-up prompts.
When to use it
- Extract line items and totals from invoice PDFs
- Summarize key points from meeting recordings
- Generate photorealistic images from text descriptions
- Transcribe lecture videos with section timestamps
- Compare two images and describe their differences
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Extract the vendor name, line items, and total from invoice.pdf
- 1Run vlmrun chat with the extraction prompt and invoice.pdf as input file
- 2Upload the PDF to the VLM Run API concurrently
- 3Orion processes the document and identifies structured fields
- 4Return extracted data as raw JSON with the --json flag
Structured JSON containing vendor name, itemized line entries, and total amount from the invoice
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.