Email safety for AI agents: scan inbound for prompt-injection/phishing + outbound for secret leaks.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add email-guard -- npx -y email-guard-mcp~/.codex/config.toml
[mcp_servers.email-guard]
command = "npx"
args = ["-y", "email-guard-mcp"]
# codex mcp add email-guard -- npx -y email-guard-mcp~/.cursor/mcp.json
{
"mcpServers": {
"email-guard": {
"command": "npx",
"args": [
"-y",
"email-guard-mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"email-guard": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"email-guard-mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"email-guard": {
"command": "npx",
"args": [
"-y",
"email-guard-mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"email-guard": {
"command": "npx",
"args": [
"-y",
"email-guard-mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"email-guard": {
"command": "npx",
"args": [
"-y",
"email-guard-mcp"
]
}
}
}