Skip to content

Configuration

All configuration is set via environment variables (loaded from .env file, supports hot-reload).

Required

VariableDescription
LLM_API_KEYLLM provider API key

Core

VariableDefaultDescription
HTTP_PORT9090HTTP server port
API_KEY(none)API authentication key (setup wizard on first run)
JWT_SECRET(none)JWT signing secret
LOG_LEVELinfoLog level (debug/info/warn/error)

LLM

VariableDefaultDescription
LLM_BASE_URLhttps://api.openai.com/v1LLM API endpoint
LLM_MODELgpt-4o-miniDefault model
LLM_FAST_MODEL(auto)Fast model (simple tasks)
LLM_SMART_MODEL(auto)Smart model (general tasks)
LLM_EXPERT_MODEL(auto)Expert model (complex reasoning)

Multiple LLM providers can be registered dynamically via the Web UI (OpenAI/Anthropic/Google/local models, etc.).

Messaging Channels

VariableDescription
TELEGRAM_BOT_TOKENTelegram bot token
FEISHU_APP_ID / FEISHU_APP_SECRETFeishu app credentials
DISCORD_BOT_TOKENDiscord bot token
SLACK_BOT_TOKENSlack bot token
WHATSAPP_TOKENWhatsApp token
WECOM_CORPID / WECOM_AGENT_ID / WECOM_SECRETWeCom credentials
DINGTALK_CLIENT_ID / DINGTALK_CLIENT_SECRETDingTalk credentials
LINE_CHANNEL_SECRET / LINE_CHANNEL_TOKENLINE credentials
KOOK_TOKENKook bot token
QQ_APP_ID / QQ_APP_SECRETQQ Official Bot credentials
SATORI_ENDPOINTSatori protocol endpoint

RAG & Embeddings

VariableDescription
EMBEDDING_MODELEmbedding model name
JINA_API_KEYJina Reranker API key
COHERE_API_KEYCohere Reranker API key

Sandbox

VariableDescription
SANDBOX_BACKENDSandbox backend: process, docker, k8s, wasm
SANDBOX_DOCKER_IMAGESandbox Docker image
E2B_API_KEYE2B Desktop sandbox API key

Hot Reload

Yunque supports config hot-reload — .env file changes are auto-loaded without restart. Implemented via ConfigLoader with support for change handlers.

© 2025 云鸢科技(青岛)有限公司 × Dream Lab