调用智谱 AI 模型做文本生成、图片生成、视频生成和多模态理解。
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 ai-xinzine --env KEY='<KEY>' -- npx -y multi-modal-mcp@latest~/.codex/config.toml
[mcp_servers.ai-xinzine]
command = "npx"
args = ["-y", "multi-modal-mcp@latest"]
env = { KEY = "<KEY>" }
# codex mcp add ai-xinzine --env KEY='<KEY>' -- npx -y multi-modal-mcp@latest~/.cursor/mcp.json
{
"mcpServers": {
"ai-xinzine": {
"command": "npx",
"args": [
"-y",
"multi-modal-mcp@latest"
],
"env": {
"KEY": "<KEY>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"ai-xinzine": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"multi-modal-mcp@latest"
],
"env": {
"KEY": "<KEY>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"ai-xinzine": {
"command": "npx",
"args": [
"-y",
"multi-modal-mcp@latest"
],
"env": {
"KEY": "<KEY>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"ai-xinzine": {
"command": "npx",
"args": [
"-y",
"multi-modal-mcp@latest"
],
"env": {
"KEY": "<KEY>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"ai-xinzine": {
"command": "npx",
"args": [
"-y",
"multi-modal-mcp@latest"
],
"env": {
"KEY": "<KEY>"
}
}
}
}需要的环境变量
KEY
把占位值换成你自己的密钥；不要把密钥提交到代码仓库。