aimond/sapiens-web3
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add personalized AI chatbot for media outlets

Browse Source 
Implement AI chatbot functionality using Replit AI Integrations and OpenAI's GPT-5 model, storing chat history in a new `chatMessages` table.

Replit-Commit-Author: Agent
Replit-Commit-Session-Id: 9a264234-c5d7-4dcc-adf3-a954b149b30d
Replit-Commit-Checkpoint-Type: full_checkpoint
Replit-Commit-Screenshot-Url: https://storage.googleapis.com/screenshot-production-us-central1/3df548ff-50ae-432f-9be4-25d34eccc983/9a264234-c5d7-4dcc-adf3-a954b149b30d/d35d7YU
This commit is contained in:
kimjaehyeon0101
2025-10-15 02:13:09 +00:00
parent fb1d150554
commit 1fe66ac046
2 changed files with 24 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
replit.md
View File

@ -32,6 +32,7 @@ The system uses a PostgreSQL database with the following core entities:
- **Comments**: User-generated content on articles
- **Community Posts**: DC Inside-style forum posts for each media outlet with views, likes, and replies
- **Community Replies**: Threaded replies to community posts
- **Chat Messages**: AI chatbot conversation history per user per media outlet
## Authentication & Authorization
- **Multi-tier Access**: Three user roles with escalating permissions
@ -47,6 +48,7 @@ The system uses a PostgreSQL database with the following core entities:
- **Content Management**: Rich article creation, editing, and deletion with media outlet administration
- **DC Inside-style Community**: Forum-style discussion boards for each media outlet with posts, replies, likes, and sorting options
- **Community Features**: Post creation, threaded replies, view counts, like system, and multiple sort options (latest, views, likes, replies)
- **AI Chatbot**: Personalized AI assistant for each media outlet that adapts its personality based on outlet category (people/topics/companies), providing contextual conversations in Korean
# External Dependencies
@ -68,6 +70,23 @@ The system uses a PostgreSQL database with the following core entities:
- **React Hook Form**: Form handling with validation
## Third-party Integrations
- **Replit AI Integrations (OpenAI)**: GPT-5 model for AI chatbot functionality via Replit-managed API (no separate API key required)
- **Kalshi API**: External prediction market data (referenced in user requirements)
- **Image Services**: Unsplash for placeholder images and media content
- **Session Storage**: connect-pg-simple for PostgreSQL session management
- **Session Storage**: connect-pg-simple for PostgreSQL session management
# Recent Changes
## January 15, 2025 - AI Chatbot Implementation
- **Feature**: Added personalized AI chatbot for each media outlet
- **Implementation**:
- Database: Created chatMessages table to store conversation history
- Backend: Implemented chatbot service using OpenAI GPT-5 via Replit AI Integrations
- Personality System: AI adapts based on outlet category:
* People: Speaks as the person themselves
* Topics: Acts as an expert on the topic
* Companies: Represents the company
- API: GET and POST endpoints for chat history and message sending
- Frontend: ChatbotModal component with scrollable chat interface
- Integration: Added Chat button to media outlet pages (authenticated users only)
- **Testing**: End-to-end test passed successfully with authentication flow and message exchange