多模型视觉理解 MCP 服务器，为不支持图片理解的 AI 编码模型提供视觉能力：分析截图、报错、UI 与文档，可接入多家主流视觉大模型。Multi-model vision MCP server that adds image understanding to AI coding models without native vision — analyze screenshots, errors, UI and documents via major vision LLM providers.
接入配置
选择客户端，复制配置，再按项目文档补齐环境变量。
claude mcp add luma --env MODEL_PROVIDER='<MODEL_PROVIDER>' -- npx -y luma-mcp~/.codex/config.toml
[mcp_servers.luma]
command = "npx"
args = ["-y", "luma-mcp"]
env = { MODEL_PROVIDER = "<MODEL_PROVIDER>" }
# codex mcp add luma --env MODEL_PROVIDER='<MODEL_PROVIDER>' -- npx -y luma-mcp~/.cursor/mcp.json
{
"mcpServers": {
"luma": {
"command": "npx",
"args": [
"-y",
"luma-mcp"
],
"env": {
"MODEL_PROVIDER": "<MODEL_PROVIDER>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"luma": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"luma-mcp"
],
"env": {
"MODEL_PROVIDER": "<MODEL_PROVIDER>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"luma": {
"command": "npx",
"args": [
"-y",
"luma-mcp"
],
"env": {
"MODEL_PROVIDER": "<MODEL_PROVIDER>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"luma": {
"command": "npx",
"args": [
"-y",
"luma-mcp"
],
"env": {
"MODEL_PROVIDER": "<MODEL_PROVIDER>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"luma": {
"command": "npx",
"args": [
"-y",
"luma-mcp"
],
"env": {
"MODEL_PROVIDER": "<MODEL_PROVIDER>"
}
}
}
}需要的环境变量
MODEL_PROVIDER
把占位值换成你自己的密钥；不要把密钥提交到代码仓库。