MovGP0

name	vision-transcribe
description	OCR/transcribe or interpret an image file via OpenAI vision (defaults to gpt-5.2). Models gpt-5.2, gpt-5.1, gpt-5-mini, gpt-4.1, gpt-4.1-mini.
metadata	short-description Transcribe/interpret an image (Markdown output).

Purpose

This skill reads a local image file (PNG/JPG/WebP/GIF) and uses an OpenAI vision-capable model to:

• OCR / transcription (verbatim text extraction)

MovGP0

param(
    [string]$Url,
    [string]$OutputPath,
    [int]$ChunkSizeMb = 1,
    [int]$MaxRetries = 16
)

$ErrorActionPreference = 'Stop'

if ($ChunkSizeMb -lt 1)