What it does. Five local CLIs that give a text-only agent eyes. They read one shared vision config (
VISION_API_KEY
/
VISION_BASE_URL
/
VISION_MODEL
/
LANG
), plus the optional Python-client settings
VISION_API_PROTOCOL
,
VISION_REASONING_EFFORT
, and
VISION_USER_AGENT
— no extra credentials.
glance
answers what something is;
ground
and
detect
answer where. You give
ground
a description of a particular thing; you give
detect
a kind and it enumerates the instances.
Highlights
- cut a box out of an image →
crop
, not Image.open(...).crop(...)
- sample a region's palette →
scripts/dominant_colors.py
- compare two images →
scripts/pixel_diff.py
- vectorize to SVG →
trace
- locate / inventory elements →
ground
/ detect
When to use it. Local vision CLIs that let a text-only model see: describe, ask about or OCR an image, and locate a target on screen for GUI automation.
Install
npx skills add Anionex/agent-vision-toolkit --skill vision-skills
Or copy the skill folder manually:
git clone https://github.com/Anionex/agent-vision-toolkit.git
cp -r agent-vision-toolkit/skills/vision-skills/ ~/.claude/skills/vision-skills
Source: Anionex/agent-vision-toolkit — 1,137 stars, MIT, last pushed 2026-08-27.
