#1010th of 42 in Agent platformsFor data and documents, data analysis
bettafish
Multi-agent public opinion analysis over social media and web search
- Stars
- 42.5k
- License
- GPL-2.0
- Last commit
- Oct 2026
- Last release
- Dec 2025
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
BettaFish (微舆) runs Query, Media and Insight agents in parallel over web search and crawled social media data. A forum mechanism with an LLM moderator steers them across rounds, and a Report agent renders an interactive HTML report, with optional PDF and Markdown export. A Flask app serves the UI, and a bundled crawler (MindSpider) fills a PostgreSQL or MySQL database.
Who it is for: Analysts and developers who want self-hosted social media sentiment reports
Ranked among open-source alternatives to Copilot Studio (#10 of 10).
Strengths
- Bundles crawler, sentiment models (BERT, GPT-2, Qwen3 LoRA) and report renderer in one Python codebase
- Forum engine lets agents exchange findings over several rounds before the report is written
- Any OpenAI-compatible LLM API, configurable per agent through a single .env file
- Each agent also runs standalone as a Streamlit app; a CLI regenerates reports from earlier runs
Weaknesses
- Needs a PostgreSQL or MySQL database, and crawling is run separately before there is data
- Main README is in Chinese; English docs live in a separate file
- Built for opinion analysis; other domains require editing agent tools and prompts
- PDF export needs extra system libraries for WeasyPrint
What it needs
- RAM ≥ 2 GB
- GPU optional
- Docker + Compose
- Needs PostgreSQL or MySQL, OpenAI-compatible LLM API, Playwright Chromium (crawler), WeasyPrint system libraries (PDF export, optional)
- Models: OpenAI-compatible API
- port 5000
Also in Agent platforms
See all 42| Rank | Project | Score |
|---|---|---|
| 1 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 81 out of 100 |
| 2 | daily_stock_analysisDaily LLM stock analysis for A-share, HK, US and other markets, pushed to chat apps | 81 out of 100 |
| 3 | resume-matcherTailors resumes and cover letters to job descriptions using local or hosted LLMs | 81 out of 100 |
| #4 | vibe-tradingFinance research agent for backtests, market data and broker connectors | 78 out of 100 |
| #5 | deer-flowAgent harness that runs sub-agents, memory and sandboxed skills for long tasks | 77 out of 100 |