MCP server that gives AI agents typed decisions instead of prose: classify, route, score, and flag with calibrated probabilities. npx -y openjev-mcp
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 openjev --env OPENJEV_API_KEY='<OPENJEV_API_KEY>' -- npx -y openjev-mcp~/.codex/config.toml
[mcp_servers.openjev]
command = "npx"
args = ["-y", "openjev-mcp"]
env = { OPENJEV_API_KEY = "<OPENJEV_API_KEY>" }
# codex mcp add openjev --env OPENJEV_API_KEY='<OPENJEV_API_KEY>' -- npx -y openjev-mcp~/.cursor/mcp.json
{
"mcpServers": {
"openjev": {
"command": "npx",
"args": [
"-y",
"openjev-mcp"
],
"env": {
"OPENJEV_API_KEY": "<OPENJEV_API_KEY>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"openjev": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"openjev-mcp"
],
"env": {
"OPENJEV_API_KEY": "<OPENJEV_API_KEY>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"openjev": {
"command": "npx",
"args": [
"-y",
"openjev-mcp"
],
"env": {
"OPENJEV_API_KEY": "<OPENJEV_API_KEY>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"openjev": {
"command": "npx",
"args": [
"-y",
"openjev-mcp"
],
"env": {
"OPENJEV_API_KEY": "<OPENJEV_API_KEY>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"openjev": {
"command": "npx",
"args": [
"-y",
"openjev-mcp"
],
"env": {
"OPENJEV_API_KEY": "<OPENJEV_API_KEY>"
}
}
}
}需要的环境变量
OPENJEV_API_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。