ReelAgent — 自研ReAct Agent驱动的短剧工厂。无LangChain，纯手搓Agent框架。剧本→成片全自动，支持shot级审核重生。
Manage your LLM prompts offline with this Tkinter and SQLite tool. Organize, search, and edit your local prompt library without accounts, servers, or web access.
Agent IA d'analyse documentaire — RAG hybride + ReAct + double check anti-hallucination. Stack Mistral / FAISS / Streamlit. Évalué sur 5 métriques RAGAS-inspired.
AI research agent that autonomously selects and calls tools (search, Wikipedia, arXiv, calculator) to answer questions with cited sources. Built with FastAPI + Groq.
Autonomous local AI agent. Ollama, FastAPI, React 19. ReAct pattern, file upload, IDE project reader, and MCP integration.
A lightweight Python AI agent implementing the ReAct (Reasoning and Acting) pattern. Powered by Claude and equipped with weather and search tools for travel planning.
AI-powered CLI that generates valid cloud-init user-data YAML from natural language using a ReAct agentic loop (plan → generate → validate → fix). Supports Ollama and Anthropic Claude backends.
A real-time physics playground simulating a robotic Virtual Hardware Abstraction Layer (V-HAL). Hot-swap between human │ control, a sub-millisecond PyTorch MLP, and a ReAct-based Ollama LLM Agent to explore latency, spatial reasoning, and │ the limits of Generative AI in robotics.
LangGraph ReAct agent with guardrails, MCP-contract tool-use, conversation memory, and a DuckDB/Parquet lakehouse — served as an async FastAPI microservice with CI/CD.
LangGraph ReAct agent with short-term (session) and long-term (Redis) memory. Graph-based Reason-Act-Observe loop with ToolNode, calculator tool, and cross-session fact recall — a direct comparison to a hand-built ReAct implementation.
🤖 Minimal CLI agent using single-turn ReAct pattern with OpenAI LLM wrapper. Intelligently selects between calculator, string operations, and direct responses. Extensible tool system with comprehensive testing.
Production-grade ReAct agent on Claude autonomously orchestrating Slack, Jira, Calendar and Email through a typed FastAPI service. Native tool use, structured outputs, multi-step reasoning, full test coverage. Built in Python with the Anthropic SDK.
To associate your repository with the react-patterns topic, visit your repo's landing page and select "manage topics."