666ghj/bettafish is a multi-agent assistant that automates monitoring and analysis of public opinion across more than 30 social media platforms. It coordinates specialized agents through a debate mechanism and runs locally as a Flask web application, requiring external OpenAI-compatible LLM APIs and PostgreSQL.
Project overview
The project implements a multi-agent system from scratch without relying on existing agent frameworks, using a Forum mechanism where agents debate under an LLM host to analyze social media comments from a single chat prompt.
Project type
AI Agent
Use cases
Search & Research · Data Analysis
Deployment
Refer to project documentation
License
GPL-2.0
Best for
Developers seeking a self-hosted, multi-agent framework for cross-platform public opinion analysis who are comfortable configuring environment variables and external LLM APIs.
Key capabilities
Implements a multi-agent system featuring Query, Media, Insight, and Report Agents from scratch without relying on existing agent frameworks.
Uses a Forum mechanism with an LLM host to enable chain-of-thought collisions and debates among different agents.
Limitations and risks
The PDF export functionality relies on WeasyPrint, which requires specific system dependencies to be installed.
Getting started
Setup is rated as medium difficulty. Users must copy .env.example to .env and configure variables, install dependencies via pip, install Playwright browser drivers, initialize a PostgreSQL database, and run the Flask application using python app.py.
Alternatives and comparisons
Provides an API for searching, scraping, and interacting with web content at scale, turning pages into Markdown or structured data.
A multi-agent LLM framework for trading that mirrors a real-world firm with specialized analysts and researchers.
An open-source harness that orchestrates sub-agents, memory, and sandboxes for complex long-horizon tasks.