Answer

Node Overview

Node name: Answer
Supported scenarios: Chatflow Agent (supported), Workflow Agent (not supported)

Feature Description

The Answer node outputs node results or the final response to the Chat system. It is commonly used to end a conversation turn or provide the final information to the upstream caller. It is the terminal node in a Chatflow Agent, responsible for formatting and outputting the final answer content.

Node Display Content

1. Basic Information Display

• Icon: node-specific icon
• Name: node name
• Output template: a preview of the template content with blank lines removed
• Note: explanatory text added by the user

2. Action Buttons

• More button: click to display the context menu content

Node Edit Panel

1. Reply Template Editing

a. Editor Features

• Variable insertion: various variables can be inserted into the template
• Text input: plain text content can be entered directly
• Variable selection: type "/" to bring up the variable value selector

b. Editing Aids

• Length counter: counts the template length; variable names count toward the template length
• Copy: click the copy button to copy the template
• Maximize: click the maximize button to expand the template view

2. Save Context

The Save Context feature is essential for context preservation. It saves the current turn's context when the Chatflow run finishes, making it available to subsequent conversation turns. This context includes all kinds of information about tool calls.

• Select the context variable to save here, e.g., LLM/context (of type array[context]) is selected in the image above
• If a Chatflow needs solid context management, you must save the context variable of the LLM node or Agent node here—otherwise, in the next conversation turn, the model will have no awareness of this turn's tool call process and intermediate results

3. Long-Term Memory

• When enabled, extracted long-term memories are displayed synchronously as the content is output
• This is not a mandatory feature; enable it as needed
• When enabled, the output variables section gains user_memory_changes (of type array[string]), which records the long-term memory changes extracted in this turn

4. Additional Data

• The Additional Data feature is used to implement certain features of the main Agent
• Agents built by regular users generally cannot use it yet; related feature tutorials will be opened up gradually

5. Output Variables

• answer: String type, the final answer content output by this node
• user_memory_changes: array[string] type, the long-term memory change records (appears only when Long-Term Memory is enabled)

6. Next Step Configuration

• The Answer node is a terminal node and does not support adding a next node

Execution Logic

The execution logic of the Answer node is implemented by the backend to ensure the answer is output correctly.

Context Menu Actions

The context menu contains the following options:

1. Change node: change the node type
2. Copy: copy the node
3. Duplicate: duplicate the node content
4. Delete: delete the node
5. Help link: jump to the help documentation

Node Count Limits

• Limit: an Agent can have multiple Answer nodes
• Purpose: can be used as a terminal node; multiple Answer nodes suit different branches—for example, after a condition check or question classification, each branch leads to its own Answer node and outputs the corresponding reply content

Test Run

• Support status: test run not supported
• Change node: changing the node type is supported
• Copy features: copy, cut, and delete operations are supported

Use Cases

1. Conversation Termination

• Serves as the final output of a conversation in a Chatflow Agent
• Presents AI-processed results to the user in a friendly way

2. Result Formatting

• Formats complex processing results into user-readable text
• Supports variable substitution to dynamically generate personalized content

3. Multilingual Support

• Supports output templates in different languages
• Output format can be adjusted based on the user's language preference

Template Writing Tips

1. Directly Reference Upstream Node Output (Most Common)

Most scenarios do not require a complex template—simply select the output variable of the LLM node or Agent node:

• In the reply template editor, type / to bring up variable selection and pick the upstream node's text output (e.g., LLM/text)
• Put only this single variable in the template; whatever the model generates is output verbatim—simple and reliable

2. Debug Output Tips

When debugging a Chatflow, you can temporarily add intermediate variables to the template to print the flow's internal state directly into the reply for inspection:

[Debug]
Classification result: {{classifier node output}}
Retrieval hits: {{knowledge retrieval output}}
---
{{LLM/text}}

• This quickly pinpoints where the problem lies: a wrong branch, empty retrieval, or abnormal model generation
• You can also take advantage of multiple Answer nodes by attaching a dedicated debug Answer node to a specific branch to output that branch's intermediate variables
• Remember to remove the debug content before publishing to avoid exposing internal information to end users

FAQ

Q: How do I insert variables?

A: You can insert variables by typing "/".

Q: How is the template length calculated?

A: The template length includes all text content and variable names; variable values do not count toward the length.

Q: Which variable types are supported?

A: Basic data types such as String, Number, Object, and Array are supported.

Q: How do I handle empty values?

A: It is recommended to use conditional checks in the template to handle possible empty values.