How to Use action_fastline_pro Action in Scenario?

The action_fastline_pro action allows using an AI agent or chatbot created in FastLinePro in a bot scenario. This allows integrating smart AI responses into the dialogue with the user.

When Needed

You need to use an AI agent to answer user questions.
You need to integrate a chatbot with a knowledge base into the scenario.
You need to automate responses to typical questions via AI.

What's Important to Know

action_fastline_pro — action for calling an AI agent from FastLinePro.
Agent name must exactly match the name specified in FastLinePro.
Conversation ID is responsible for saving dialogue context.
Vision file allows the agent to process images and files.

Prerequisites

You are logged in with integrator or administrator rights.
There is a created AI agent or chatbot in FastLinePro in the system.
You have access to the scenario constructor.

Step-by-Step Instructions

In the scenario constructor, add an "Action" block in the needed place in the scenario.
In the "Action" block settings, select the action_fastline_pro function.
In the configuration field (JSON), enter parameters:

{
  "agent_name": "Your company agent",
  "user_input": "{{question}}",
  "vision_file": true,
  "conversation_id": "fl_client_id",
  "save_response": "client_answer",
  "inputs": {
    "first_name": "Bohdan"
  }
}

Action Parameters

agent_name — name of the AI agent you created and connected in the interface. Must exactly match the name specified in Fast Line Pro.
user_input — user question or query. Usually this variable is filled in the previous step of the scenario, for example: {{question}}.
vision_file — boolean parameter that determines whether the agent has access to files, images, and other attachments:
- true — agent sees files
- false — agent works only with text
conversation_id — variable responsible for saving dialogue context. If you pass an empty value — the agent will treat the query as the start of a new dialogue. After the first response, it will generate a conversation_id itself, which should be saved and passed in subsequent queries.
save_response — name of the variable where the agent's response will be saved. For example: client_answer.
inputs — object containing all variables you want to pass to the agent (must be previously created in the agent's "Variables" block).

Save changes in the action and scenario.

Usage Example

Scenario:

"Wait for Response" block → user enters a question, saved in {{question}}
"Action" block with action_fastline_pro → agent responds, saved in {{client_answer}}
"Message" block → displays {{client_answer}}

What Happens After

After executing the action, the AI agent processes the user's query and generates a response based on its instructions and knowledge base (if connected). The response is saved to the variable specified in save_response and can be used in subsequent scenario blocks.

Passing Context Between Agents

The feature is enabled per instance by the ff_fastline_pro_context_summary feature flag. Until it is turned on, the settings card is not shown.

When one scenario uses several action_fastline_pro actions with different agents, you can pass a short summary of the earlier conversation to the next agent.

Open Settings → Bots and select the bot.
In the card Context between Fastline Pro agents, turn on Pass context.
Select the Summarizer agent (required) and, if needed, add an Extra prompt (up to 4000 characters).
Choose the Scope: all scenario agents or selected agents.
Optionally list Constants to pass - names of scenario constants (without curly braces) whose values are added to the context.
Click Save.

What the next agent receives: when the dialog moves from agent A to agent B, the request to B starts with a block [Контекст діалогу від попередніх агентів] (the block labels are always in Ukrainian) containing the summary and the values of the chosen constants, followed by the original message. If the summary is not empty, it is also passed to the agent as the input dialog_context_summary and saved to the scenario constant fastline_context_summary, so you can use {{fastline_context_summary}} in the fields of the following scenario nodes.

The summary is built only when the dialog moves to a different agent, so consecutive messages to the same agent get no extra delay. Each summary is a separate call to the summarizer agent and counts toward AI usage the same way as a regular agent answer. If the summarizer does not answer within 20 seconds or returns an error, the next agent answers as usual, without a new summary.

How to Verify It Worked

Check that the agent name in agent_name exactly matches the name in FastLinePro.
Make sure all parameters are specified correctly.
Test the scenario and check that the agent responds correctly.
Check that the response is saved to the variable from save_response.