Free Vision Skill
niyongsheng/free-vision-skill · model-adapters
Fully-local image understanding (OCR / table extraction / description) via macOS Vision.
★ 5GitHub stars
0Forks
2026-08-27Last updated
TypeScriptLanguage
MITLicense
Key features
- view_image — describe image content (scene, people, QR, composition)
- Requires macOS 11+ & Xcode Command Line Tools
- layout=true for table structure + coordinates
- First run compiles 5–10s, cached afterwards
Install command
dsh plugin --profile web add github:niyongsheng/free-vision-skill