Lightweight AI agent orchestrator with built-in Architect AI. Multi-team support, MCP, web dashboard.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add zoogent --env ZOOGENT_API_KEY='<ZOOGENT_API_KEY>' -- npx -y zoogent~/.codex/config.toml
[mcp_servers.zoogent]
command = "npx"
args = ["-y", "zoogent"]
env = { ZOOGENT_API_KEY = "<ZOOGENT_API_KEY>" }
# codex mcp add zoogent --env ZOOGENT_API_KEY='<ZOOGENT_API_KEY>' -- npx -y zoogent~/.cursor/mcp.json
{
"mcpServers": {
"zoogent": {
"command": "npx",
"args": [
"-y",
"zoogent"
],
"env": {
"ZOOGENT_API_KEY": "<ZOOGENT_API_KEY>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"zoogent": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"zoogent"
],
"env": {
"ZOOGENT_API_KEY": "<ZOOGENT_API_KEY>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"zoogent": {
"command": "npx",
"args": [
"-y",
"zoogent"
],
"env": {
"ZOOGENT_API_KEY": "<ZOOGENT_API_KEY>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"zoogent": {
"command": "npx",
"args": [
"-y",
"zoogent"
],
"env": {
"ZOOGENT_API_KEY": "<ZOOGENT_API_KEY>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"zoogent": {
"command": "npx",
"args": [
"-y",
"zoogent"
],
"env": {
"ZOOGENT_API_KEY": "<ZOOGENT_API_KEY>"
}
}
}
}Environment variables
ZOOGENT_API_KEY
Replace the placeholders with your own credentials and keep them out of version control.