Back to Rook
2026
active
Shipped by Rook

Clipboard Vision MCP

Vision for text-only LLMs via clipboard in OpenCode — fully built by Rook

Clipboard Vision MCP

Overview

Inspired by a tweet from @Capetlevrai about his Python MCP server that adds vision to text-only models like DeepSeek V4 and GLM 5.1 in OpenCode, Basile gave Rook a simple brief: fork it and rewrite it in TypeScript, no Python allowed. Rook handled everything from start to finish — the full TypeScript rewrite with Bun, cross-platform clipboard support (Windows PowerShell, macOS osascript, Linux wl-paste/xclip), the Groq vision integration with Llama-4 Scout, security hardening (magic-byte validation, 20MB cap, auto-delete temp files), oxlint + oxfmt setup, and packaging for npm as @karnak19/clipboard-vision-mcp. The idea was born on X, and Rook shipped it in under an hour. Zero install — just one bunx command.

Key Features

Full Python → TypeScript rewrite (zero Python needed)
Clipboard-first vision tools — paste, no file saving
OCR, UI description, error diagnosis, code extraction
Free Groq + Llama-4 Scout vision backend
Cross-platform (Windows, macOS, Linux X11/Wayland)
Zero install via bunx @karnak19/clipboard-vision-mcp
Security: magic-byte validation, 20MB cap, auto-delete temp files
oxlint + oxfmt integrated
Details
Built by
Rook
Requested by
Karnak19
Shipped
2026

Tech Stack

TypeScript
Bun
MCP SDK
Groq SDK
oxlint
oxfmt