This document gives a step-by-step breakdown of how prompts flow through the system:
  • From frontend trigger to backend Python handlers
  • Includes hooks, handlers, payload formats, and utility functions
  • Ideal for developers working on prompt submission, socket handling, and agent-based AI routing

​
Core Functions

​
handleAIApiType(fileMetadata)

Determines which backend API handles the request based on file metadata. Returns: String key (doc, custom_gpt, canvas) for backend routing

​
handleSubmitPrompt(refine = false)

Triggers the complete prompt submission lifecycle. Behavior:
  • First message: Creates new chat and adds user as member
  • Always disables chat input until response received

​
enterNewPrompt()

Creates message entry in MongoDB before backend submission. Usage: Must be called before sending to backend

​
Payload Structure

Standard payload format across all LLM APIs: 
{
  query: string,
  messageId: string,
  modelId: string,
  chatId: string,
  model_name: string,
  msgCredit: number
}

​
Backend API Functions

​
General Chat Functions

  • getPerplexityResponse(payload) - General LLM processing
  • getAINormatChatResponse(payload) - Image uploads or no prompt selected
  • getAIDocResponse(payload) - Document-based chats
  • setChatTitleByAI(params) - Auto-generate chat titles

​
Agent Functions

  • getAICustomGPTResponse(payload) - Custom GPT agents (uses agent.model_name)
  • getAIProAgentChatResponse(payload) - QA and Web Proposal agents
  • getSalesCallResponse(payload) - Sales call analysis

​
Specialized Functions

  • chatCanvasAiResponse(payload) - Selection-based canvas responses
  • getSeoKeyWords(payload) - SEO keyword generation

​
React Hooks

​
Core Hooks

  • useConversation() - Central conversation management, LLM streaming, prompt submission
  • useThunderBoltPopup() - File uploads, agent selection, prompt selection
  • useMediaUpload() - Media uploads for canvas responses

​
Data Hooks

  • useCustomGpt() - Fetch available custom GPT agents
  • usePrompt() - Retrieve saved prompts
  • useBrainDocs() - Load uploaded documents

​
Utility Hooks

  • useIntersectionObserver() - Infinite scroll for lists
  • useDebounce() - Debounce user input
  • useServerAction() - Server-side actions in Next.js

​
Utility Functions

​
State Management

  • handleModelSelectionUrl() - Sync model state with URL
  • handleProAgentUrlState() - Sync Pro Agent state with URL
  • handleNewChatClick() - Start new chat session

​
UI Control

  • blockProAgentAction() - Disable input during Pro Agent response

​
API Helpers

  • commonApi() - Generic client-side API calls
  • serverApi() - Server-specific API wrapper
  • ValidationError - Display Yup validation errors

​
UI Components

​
Lists and Modals

  • CommonList - Combined private/shared document list
  • RenderModalList - Model selection dropdown

​
Common Parameters

​
Standard Parameters

  • custom_gpt_id - Agent identifier
  • companyId - Company identifier
  • provider - LLM provider
  • code - Model code

​
Canvas-specific Parameters

  • currentMessageId - Message identifier
  • startIndex - Selection start
  • endIndex - Selection end

​
Agent-specific Parameters

  • agent_extra_info - Additional agent data

​
Troubleshooting

​
Prompt Not Reaching Backend

  1. Verify handleSubmitPrompt() is triggered
  2. Ensure enterNewPrompt() runs before submission
  3. Check handleAIApiType() returns correct type
  4. Confirm socket connection is active

​
Custom GPT Issues

  • Verify getAICustomGPTResponse() uses agent.model_name
  • Check payload structure matches expected schema

​
Pro Agent Input Issues

  • Ensure blockProAgentAction() is called before getAIProAgentChatResponse()