Description

Configuration for a single LangChain4j agent.

Usages

Configuration options

KeyKindTypeDefault ValueDescription
asyncVALUEBoolean If true, the agent will be invoked in an asynchronous manner, allowing the workflow to continue without waiting for the agent's result
chat-memoryVALUEString Name of the dev.langchain4j.memory.ChatMemory service to use for this agent
chat-memory-providerVALUEString Name of the dev.langchain4j.memory.chat.ChatMemoryProvider service to use for this agent
chat-modelVALUEString Name of the dev.langchain4j.model.chat.ChatModel service to use for this agent
content-retrieverVALUEString Name of the dev.langchain4j.rag.content.retriever.ContentRetriever service to use for this agent
descriptionVALUEString Description of the agent
enabledVALUEBooleantrueIf set to false, agent will not be available even if configured
execute-tools-concurrentlyVALUEBoolean If true, the agent's tools can be invoked in a concurrent manner
input-guardrailsLISTClass Input guardrail classes to apply to the agent
mcp-clientsLISTString Names of dev.langchain4j.mcp.client.McpClient services to use for MCP-backed tools
nameVALUEString Agent identifier used to label the agent in workflows and/or agent registries
output-guardrailsLISTClass Output guardrail classes to apply to the agent
output-keyVALUEString Key of the output variable that will be used to store the result of the agent's invocation
retrieval-augmentorVALUEString Name of the dev.langchain4j.rag.RetrievalAugmentor service to use for this agent
tool-providerVALUEString Name of the dev.langchain4j.service.tool.ToolProvider service to use for this agent
toolsLISTClass Tool service classes to register with the agent

See the manifest for all available types.