Generates all 5 meeting prep sections in parallel using LLM with structured JSON output. Uses tiered models for speed/quality balance.
Bearer token from your account (https://agent.ai/user/integrations#api)
The processed contact research results.
Target company information.
The meeting classification result.
The processed calendar event data.
Variable name to store generated sections.
^[a-zA-Z][a-zA-Z0-9_]*$Topic signals from classification. Usually {{meeting_classification.topic_signals}}.
Relationship analysis results.
User context for personalization.
Which sections to generate. Options: overview, attendees, company, strategy, goals.
Model for simpler sections (overview, company, goals).
gpt-5-mini, gpt-5, gpt-4o-mini, gpt-4o, claude-haiku-4-5 Model for complex sections (attendees, strategy).
gpt-5, gpt-5-mini, claude-sonnet-4-5, gpt-4o, gpt-4o-mini