ningyougan/deepseek-vision-mcp
Skill Claude CodeCodex
A visual design reader that turns UI screenshots, Figma exports, prototypes, or image links into written layout details and front-end code.
0 25d ago A 80 tokens
original MIT
ningyougan/deepseek-vision-mcp
MCP server Claude CodeCodexCursor +2
MCP server for visual image recognition using OpenAI-compatible vision models. Zero-dependency single-file build, configurable via MCP client environment variables (url/apiKey/model). Runs locally from the deepseek-vision-mcp npm package.
0 25d ago A
tokens not measured
original MIT