GPT Researcher is an autonomous research agent that produces comprehensive, cited reports from web and local sources. In production, it operates as a full-stack web application: a FastAPI backend (main.py, backend/server/app.py) exposes REST and WebSocket endpoints for initiating research, streaming progress, and retrieving reports. A Next.js frontend (frontend/nextjs/) provides the user interface, and a lightweight vanilla JS frontend (frontend/scripts.js) is also available. The core Python package (gpt_researcher/) can be used programmatically via the GPTResearcher class or as a CLI tool (cli.py). Architecture follows a planner-executor pattern: given a research task, the system generates sub-queries, executes them in parallel across 20+ search engine wrappers (Tavily, Google, Bing, Brave, DuckDuckGo, ArXiv, PubMed, etc.), scrapes the resulting URLs using multiple backends (BeautifulSoup, browser automation, PyMuPDF, FireCrawl), filters and compresses the context using embedding or keyword-based methods, and finally generates a structured report with citations. The system supports multiple report types (summary, detailed, deep research) and multi-agent orchestrations (LangGraph, AG2) for more complex workflows. Key data flows: user query to sub-query generation (LLM) to parallel search across configured retrievers to URL scraping to context compression to source curation to report generation (LLM) to output as markdown/PDF/DOCX. Real-time progress is streamed via WebSocket. External dependencies include LLM providers (OpenAI, Anthropic, Google, Ollama, etc.), search APIs (Tavily, SerpAPI, etc.), and optional services (MCP servers, vector stores, image generation APIs). The codebase demonstrates strong engineering practices: extensive test suite (200+ tests), plugin architecture for retrievers, configurable context filters, cost tracking, and evaluation frameworks for quality and hallucination. However, it carries maintenance risks from its large dependency surface, multiple parallel agent implementations, and reliance on rapidly-changing external APIs.
BettaFish (åūŪč) is a multi-agent public opinion analysis system that automates the end-to-end process of gathering, analyzing, and reporting on public sentiment across 30+ Chinese social media platforms. The system is built around four specialized agents: QueryEngine (news breadth search via Tavily), MediaEngine (multimodal search via Bocha), InsightEngine (database mining via SQLAlchemy), and ReportEngine (template-based report generation with intermediate representation and HTML/PDF/Markdown rendering). A ForumEngine coordinates agent collaboration through log monitoring and LLM-generated moderator comments. The architecture is a monolithic Flask application that manages multiple Streamlit sub-applications and CLI tools, with a separate MindSpider crawler subsystem for data collection. While primarily an application, it includes significant data/research components such as sentiment analysis model training scripts (BERT, GPT-2, Qwen3, traditional ML), a Weibo sentiment dataset, and data pipeline scripts. The system relies heavily on external LLM APIs (OpenAI-compatible), search APIs (Tavily, Bocha, Anspire), and system dependencies (Playwright, WeasyPrint). The codebase demonstrates solid engineering with reusable patterns like robust JSON parsing, chart validation/repair, and a hardened GitHub Actions star-history script, but suffers from significant code duplication across engines and tight coupling to external APIs.
DeepReviewer-v2 is a backend-only application designed for automated academic paper review. It processes PDF papers through a pipeline involving MinerU for markdown conversion, an LLM-based review agent with tool usage (e.g., paper search, annotation), and PDF report generation with annotated source overlays. The system operates asynchronously with job persistence, featuring a CLI for job submission, status tracking, and result retrieval. Optional integration with PASA (via vLLM) enables external paper search. The architecture is modular, separating concerns into adapters, tools, state management, and report generation components, all coordinated via an async runner.
FinClaw is an open-source, finance-domain-specific autonomous AI agent execution framework, primarily targeting Chinese A-share markets and institutional banking/insurance/trust workflows. The repository contains over 1000 modular 'skills' - standalone CLI scripts that perform market data retrieval (via AkShare, Tushare, Yahoo Finance), risk scoring (keyword-based sentiment, threshold alerts, Brinson attribution), compliance checks (sensitive phrase detection, clause conflict analysis), and report generation (Markdown/JSON outputs). Each skill is self-contained with its own entry point and dependencies, designed to be orchestrated by the FinClaw Agent OS. The repository is structured as a flat collection of independent skills under `skills/`, with significant duplication of identical scripts (e.g., `analyze.py` templates, `keyword_retrieval.py`, `pack.py`) across multiple skill directories. This copy-paste pattern reduces maintainability but allows rapid creation of domain-specific assistants without shared library dependencies. Notable well-crafted components include the cn-stock-data adapter layer (abstracting multiple Chinese data sources) and the async MCP connection factory for skill networking. The codebase is heavily biased toward Chinese financial regulations and terminology, with skills for bank loan pre-screening, fund portfolio diagnostic, insurance underwriting, and securities trading strategy backtesting. Many scripts are placeholders or demos with hardcoded positive/negative keyword lists and no real AI/ML logic, relying instead on rule-based scoring. The repository is actively maintained (last commit 2026-05-13) but suffers from medium technical drift due to duplicated boilerplate, hardcoded API tokens, and fragile web scraping patterns.