Skip to content
Nappai Documentation

BasicLangchainV1.0 Agent

BasicLangchainV1.0 Agent answers questions with the AI model you connect and uses the available tools when it needs them. It has few settings and, unlike Basic Agent, it keeps the conversation in the memory component you choose.

For most assistants, Agent is a better choice: it offers more options and keeps the conversation without you having to connect anything.

When to use it

  • A simple assistant that must keep the conversation in your own database.
  • An assistant that remembers user preferences thanks to long-term memory.
  • Giving another agent a simple question-and-answer tool.

Configuration

ConnectionWhat to connect
ModelThe AI model that answers. It is needed for the agent to work.
ToolsOptional tools the agent can use.
MemoryA memory component such as Postgres Memory, Redis Memory or Mem0 Cloud Memory. Without a memory connected, the agent does not remember previous messages. If the component includes long-term memory, the agent can also save and search important details the user shares.
FieldWhat it does
InputThe user’s question, usually from Chat Input.
System PromptThe agent’s instructions. By default they ask it to answer clearly and concisely and to use the tools when needed.
Agent NameThe agent’s name when used as a tool. Default: BasicQAAgent.
Agent DescriptionWhat the agent does when used as a tool. Another agent reads it to decide when to use it.

Output

  • Response: the answer as a message. Connect it to Chat Output.
  • Agent: the configured agent, so another agent can delegate to it.
  • Tool: the agent turned into a tool, which receives a question and returns the answer. It uses Agent Name and Agent Description as the tool’s name and description.

Example

The dental clinic Sonrisa Clara wants an assistant that remembers each patient. It connects Chat Input to Input, an AI model to Model and a Postgres Memory component with its database to Memory. In System Prompt it describes the clinic’s treatments and opening hours. When a patient writes again in the same conversation, the assistant takes into account what they said before.