Definition
An Agent Node is an "intelligent decision-making + tool invocation" component designed to enable a Large Language Model (LLM) within a workflow to autonomously select and utilize tools at runtime based on defined strategies, facilitating multi-step reasoning or operations.
- By mounting a specific "Agent Strategy", the model can leverage contextual instructions, tool information, and its own reasoning results to determine in each execution round whether and how to employ a tool (such as an API, database query, or plugin function). It then incorporates the data returned from these tools into the next round of reasoning or output, allowing for a more flexible processing flow.
- Compared to LLM nodes that utilize simple prompts, an Agent Node can engage in multiple cycles of "thinking and executing" according to the strategy, making it suitable for scenarios that require external data or executable actions.
Configuring an Agent Node
Configure the Agent Node
- 1. Open the Workflow Editor
- Navigate to the workflow interface you need to edit within the GoInsight.ai workspace.
- 2. Drag and Drop the Node
- Locate the "Agent" node in the left or top component panel on the canvas. Drag it to the appropriate position in the visual flowchart, connecting it with the preceding and subsequent nodes.
- 3. Select an Agent Strategy
- In the configuration panel of the Agent Node, specify an "Agent Strategy" that defines how the model will perform multi-step reasoning, invoke tools, and produce results.
- GoInsight.AI provides two default sources of Agent Strategies:
- Official System Strategies: Developed and embedded by the GoInsight.ai official team, suitable for common multi-step reasoning or tool-use scenarios.
- User-Defined Strategies (Plugins): Enterprise teams can create and publish their own Agent Strategies through the plugin module.
Configure Node Parameters
Depending on the selected Agent Strategy, the Agent Node will display various parameters or functional options. Here are some common configurations:
- 1. Model
- Specify the Large Language Model used to drive the Agent (e.g., a model instance you have already configured in GoInsight.AI).
- 2. Tools
- Add or manage external capabilities that the Agent can invoke in the "Tools" section (such as HTTP requests, database queries, search plugins, etc.).
- If certain tools or plugins are already installed, click "+" to select tools to add, thereby providing the Agent with operational interfaces.
- 3. Instruction
- Used to inform the Agent of the context, objectives, or constraints.
- You can include "Role Setting," "Task Requirements," or "Business Context" in this section, ensuring the Agent adheres to these guidelines during multi-step reasoning.
- 4. Query
- Typically corresponds to user input or requirement text supplied from upstream nodes. The Agent will perform reasoning and generate action plans or tool invocation instructions by combining the Query and Instruction.
- 5. Max Iterations
- To prevent the Agent from looping indefinitely in multi-step reasoning, you can set a maximum execution count (iteration limit).
- If the Agent exceeds this count without concluding, it will automatically stop to avoid infinite invocation rounds.
- 6. Output Variable
- Define the final output data structure of the Agent Node, enabling results to be passed to subsequent nodes for further processing.
Memory Functionality
In the configuration panel, the Agent Node offers a one-click "Memory" toggle. When memory is enabled, the Agent can retain context during multi-turn dialogues or repeated command invocations, producing coherent responses.
- Memory Window: Use a slider or numerical input to set the "Memory Window Size," indicating how many historical interactions the Agent can refer to.
- Context Coherence: Once memory is activated, the Agent can accurately interpret pronouns and unspecified field names mentioned later by the user by referencing prior content.
- Accuracy and Performance: A larger memory window may increase the model's operational costs, thus requiring a balance between actual needs and invocation expenses.
Developing and Managing Agent Strategy Plugins
If the built-in system strategies do not fulfill specific customization requirements, the technical team can utilize GoInsight.AI's plugin mechanism to create and publish custom Agent Strategy plugins.
Summary
- Agent Nodes empower large language models within workflow executions with "multi-step decision-making" and "self-service tool selection" capabilities, effectively addressing complex scenarios that simple dialog nodes cannot handle.
- You can flexibly combine models, tools, and strategies during node configuration and enhance contextual coherence and process transparency through memory and logging functions.
- If the default strategies do not meet business requirements, custom strategies can be created in the plugin module to develop intelligent workflows better tailored to specific enterprise or scenario needs.