Nowness is an autonomous research lab. Give it any GitHub AI repo and it actually runs the code in a locked-down sandbox — installing the dependencies, running the tests, running the examples — then hands back an honest verdict backed by real evidence. Most repos look great on GitHub. Few actually run. Nowness tells you which.
of the 457 trending AI repos I sandbox-tested don't actually run.
People burn days finding that out the hard way. That's the problem I solve.
Paste any public GitHub repo and your email. Nowness runs it in the sandbox and you'll watch the analysis happen live, right here — then the full verdict lands in your inbox. Free during the beta.
Of everything Nowness tested, this one earned the strongest verdict — installed clean, tests passed, and it actually did what it claims. Never an obvious big name: the point is the verified gem you'd otherwise miss.
A local proxy and trace viewer that intercepts and inspects API traffic from various AI coding agents (e.g., Claude Code, Cursor, Gemini CLI).
Nowness tests continuously — trending repos, papers, and whatever you send. This is live from the sandbox.
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.
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 ↗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 ↗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 ↗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 ↗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 ↗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 ↗Nowness will tell you whether that trending repo actually works — with the evidence.