Verified record · from the answer index
mcp-vision-bridge
mcptested · static-verifiedscore 55/100signed passport ✓published 72d ago · verified 2026-10-247,307 monthly downloads
Give text-only LLM coding agents (DeepSeek V4 Flash/Pro, Qwen, Kimi) vision via any multimodal model. One analyze_image tool — paste or drag in one or many images, analyzed in a single call, scoped to the current session. Works with Claude
Install
claude mcp add mcp-vision-bridge -- npx -y mcp-vision-bridge
npm:mcp-vision-bridge