Let your text-only coding model read images.
What it does
Converts images into text descriptions so a text-only coding model can continue its work.
When to use it
Use when the main model cannot understand images and the user supplies a screenshot, describes a visual problem, or the agent encounters image files.
How to use it
Give it images and an optional question; it returns text descriptions that the main model uses to continue the user's task.
What you provide
- Your files
Uses
Good to know
- cannot be undone: config.json
- acts without asking
- works out of sight
What you need · 4
Requires Node.js 18 or later; the recognition script uses only built-in modules.
Reads a configured model's literal API key from the configuration file, taking precedence over environment-based alternatives.
Access · 7
This skill
config.json
Saves and synchronizes model credentials (irreversible)
OpenAI-compatible APIs
Sends images to vision APIs