Action fastline_summarize
System ID:
action_fastline_summarize
Description
Fast Line: takes the latest dialog messages, filters out empty ones and those with URLs, and generates a summary from your prompt. The result is stored in a constant; if chat_room_id is set, the summary may also appear as a system message in the chat.
Events (scenario branches)
| Event | When it occurs |
|---|---|
| success | Summary generated and stored |
| no_prompt | prompt not set |
| chat_room_not_found | Chat room with the given ID not found |
| chat_user_not_found | Chat user not found |
| no_dialog_logs | No dialog logs |
| no_meaningful_content | All messages are empty or contain only links |
| error | AI generation error or other failure |
Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
prompt | string | yes | — | Summary prompt. Supports placeholders |
chat_room_id | number | no | from OP state | Chat room ID; otherwise — current chat context (without saving to OP) |
messages_count | number | no | 5 | How many latest messages to analyze |
provider | string | no | null | AI provider |
model | string | no | null | AI model |
apiKey | string | no | null | Key (also openai_token) |
customConfig | object | no | null | Additional AI settings |
temperature | number | no | 0.2 | Temperature (lower — more focused summary) |
max_tokens | number | no | 1000 | Token limit |
save_response | string | no | "summary" | Constant for the summary |
Configuration example
{
"prompt": "Briefly summarize the conversation for the next operator: main issue, customer request, current status.",
"chat_room_id": "{{chat_room_id}}",
"messages_count": 15,
"temperature": 0.3,
"save_response": "handoff_summary"
}