Running in under a minute
Paste one command into your AI chat ... or run it yourself. It detects supported clients, installs a private version-pinned runtime, writes the MCP configuration, and verifies all three tools.
$ npx --yes omni-vision-pro@latest setup --yes
Install once, from anywhere
Paste this into Codex, OpenCode, Claude Code or Cursor chat ... or your terminal:
$ npx --yes omni-vision-pro@latest setup --yes
Setup is one-time. Adding an API key later never touches your client config.
Add AI vision ... optional
Open PowerShell (or your terminal) yourself and run this command ... a friendly menu appears where you pick Gemini or OpenAI and paste your key into a hidden prompt:
$ npx --yes omni-vision-pro@latest provider
You run this yourself ... it is not done by asking your AI. Saving a key selects that provider immediately.
Already have a key copied? Run this instead ... it never touches shell history or chat text:
$ npx --yes omni-vision-pro@latest provider set --from-clipboard
Then restart or reload your AI client. Setup stays one-time ... adding a key never changes your client config.
Restart, reload & ask
Reload your AI client, then just say ...
“Use omni-vision-pro to analyze this screenshot. Return the visible text, UI hierarchy, layout, colors, interactions, accessibility issues, and reusable components.”
With no cloud key, images are processed locally with Tesseract OCR ... free and private.
Command reference
Full control from the terminal
Every command in full, ready to run in PowerShell or your terminal ... click any command to copy it. Provider settings and client setup are separate ... install once, then switch providers whenever you like.
$ npx --yes omni-vision-pro@latest setup --yes
One-click install. Detects clients, installs a pinned runtime, writes configs and verifies all three tools.
$ npx --yes omni-vision-pro@latest provider
Provider menu. Pick Gemini or OpenAI and paste your key into a hidden prompt.
$ npx --yes omni-vision-pro@latest provider status
Check the active mode. Shows which provider is in use and which keys are configured ... without ever revealing the keys.
$ npx --yes omni-vision-pro@latest provider on
Turn AI vision on. Uses any configured provider ... no need to paste keys again.
$ npx --yes omni-vision-pro@latest provider off
Use local OCR only. AI is turned off, but your saved keys are kept for later.
$ npx --yes omni-vision-pro@latest provider use gemini
Prefer Gemini. Switch anytime ... also works with use openai or use auto.
$ npx --yes omni-vision-pro@latest provider set gemini
Add or replace a key. Hidden terminal prompt ... works with set openai and replace too.
$ npx --yes omni-vision-pro@latest provider set --from-clipboard
Clipboard import. Copy your key first, then run this ... it never enters shell history or chat text.
$ npx --yes omni-vision-pro@latest provider remove gemini
Delete a saved key. From Omni's private settings ... also works with remove openai or remove all.
$ npx --yes omni-vision-pro@latest doctor
Health check. Verifies your setup and client entries ... doctor cursor checks Cursor specifically.
$ npx --yes omni-vision-pro@latest uninstall --all
Clean removal. Removes installer-managed entries only ... keeps your private .env so reinstalling never loses keys.
$ npx --yes omni-vision-pro@latest serve
Launch the MCP server. The stdio entry point that every client runs behind the scenes.
One command. All clients.
Skip the manual work ... let the installer detect, configure and verify everything for you.