Every card below was actually executed by the lab — under-the-radar repos that installed clean and did what they claim, verified in the sandbox, not guessed from the README. From 948 repos tested so far.
✓ PRODUCTION-READY
Smart-Thinking MCP Server
A Model Context Protocol (MCP) server that provides local, graph-based multi-step reasoning and memory management.
Sandbox found The sandbox successfully installed dependencies (npm_install_exit: 0) and all 20 test suites passed (npm_test_exit: 0).
github.com/Leghis/Smart-Thinking ↗
✓ USABLE
Gemini Fullstack LangGraph Quickstart
A full-stack application demonstrating a research-oriented AI agent built with LangGraph and Google Gemini.
Sandbox found The project contains a complete codebase with a frontend, backend, and deployment configurations; while the 'make test' command failed (likely due to missing test suite definitions), the project structure and documentati
github.com/google-gemini/gemini-fullstack-langgraph-quickstart ↗
✓ USABLE
Fooocus
Fooocus is a streamlined interface and implementation for Stable Diffusion image generation.
Sandbox found Static: The project has a complete file structure, including Dockerfiles, configuration files, and a comprehensive set of Python scripts, indicating a mature and released application.
github.com/lllyasviel/Fooocus ↗
✓ USABLE
Video-o3
Video-o3 is a framework for long video multi-hop reasoning that mimics human behavior by actively exploring video content.
Sandbox found The sandbox's memory cap killed the build (install_exit: 137), but the project is a complete and released research framework with training/evaluation code, datasets, and checkpoints.
github.com/MCG-NJU/Video-o3 ↗
✓ USABLE
Deep RAG
Deep RAG is an advanced Retrieval-Augmented Generation system that enhances AI understanding by preserving file structures and generating a 'knowledge map' instead of jus.
Sandbox found The project has a complete structure including a backend, frontend, and documentation, and successfully installed dependencies (install_exit: 0), though the demo_exit: 1 indicates a configuration or environment issue wit
github.com/boluo2077/deep-rag ↗
✓ USABLE
Vector Graph RAG
A Graph RAG (Retrieval-Augmented Generation) implementation that enables multi-hop reasoning using only vector search.
Sandbox found The library is published on PyPI with a clear structure and includes multiple provider support; while the sandbox's demo_exit was 1 due to a missing API key, the package successfully imported and is designed for producti
github.com/zilliztech/vector-graph-rag ↗