vision

vision is an Agent task delegation & fan-out skill for AI agents, published by WeZZard in the opencode-vision repository on GitHub. Skillspace rates vision D (68/100) and it passed the safety scan; the rating is based on static analysis only.

You **MUST** use the vision skill when when the your model is text-only (e.g. glm-5.2, deepseek-v4-pro) AND: (1) the user's message contains images; (2) OR the user's message contains URLs or paths to images; (3) OR the user asks to visually verify/check something ("visually verify", "screenshot shows", "centered/visible/hidden", "looks right", "matches the design"); (5) OR a tool result contains an image attachment the current model cannot see (attachments[].mime = "image/png", url = "data:image/png;base64,..."); (4) OR you think it is necessary to READ any visual contents; Triggers on screenshots from chrome-devtools_take_screenshot, playwright_browser_take_screenshot, cua-driver_get_window_state/zoom/take_screenshot and cannot see images itself. Extracts the visual intent from context, designs a prompt-local JSON response template for that specific task, delegates, and parses the returned JSON.

Grade

D

Quality

68 / 100

Safety

Passed

GitHub stars

22

Unproven: rated by static analysis only, not yet compared head-to-head.

This rating describes the version of the repository we crawled; it may have changed since, so go by its current files.

Skillspace indexes and rates this skill; it does not host or endorse it. Run third-party skills at your own risk.