Plugin Claude Code
Vision for text-only coding agents: analyzeimage MCP tool + vision skill + auto-loop hook (captures pasted images from the session transcript). Bring your own multimodal model (mimo, Claude, Gemini, GPT-4o).
Plugin Claude Code
Give your text-only coding agent (DeepSeek V4 Flash, Qwen, Kimi) vision. Adds an analyzeimage MCP tool that routes images through any multimodal model (mimo, Claude, Gemini, GPT-4o) and returns a detailed text description. Paste or drag in one or many images — analyzed in a single call, scoped to the current session.
Hook
Runs when you submit a prompt, before the agent sees it, executing vision-clipboard.sh via bash. From KuaaMU/mcp-vision-bridge.
Skill Claude CodeCodex
Give the agent vision when the model itself is text-only (no image input). Use this skill whenever the user asks you to look at, analyze, describe, read, or OCR any image — a screenshot, a photo, a diagram, a UI, a chart, an error dialog, a pasted picture, or anything visual. Also triggers when the user pastes an…