Introduction

VoidAccess, an open-source dark web threat intelligence platform that has garnered significant attention on Hacker News, directly challenges commercial intelligence services that easily cost tens of thousands of dollars, allowing security professionals to build their own AI-automated analysis systems. In today's "Open Source Daily", I will also bring you two other hardcore tools built with Rust: Fornjot, which turns 3D CAD into coding, and Gyroflow, a video stabilization tool used even by professional drone directors, taking you to solve real-world engineering pain points with open-source technology.

Today's Open Source Picks

1. VoidAccess: A Self-Hosted, AI-Driven Dark Web OSINT Threat Intelligence Platform

• What it is: It solves the pain point where enterprises or security researchers need to monitor dark web leaks, but commercial threat intelligence tools cost tens of thousands of dollars annually and do not allow full local control over data. It provides a self-hosted solution, enabling researchers to safely collect and analyze dark web threat intelligence in an isolated environment.
• Recent Buzz: It received 188 points on Hacker News (discussion thread: https://news.ycombinator.com/item?id=41824147), currently has accumulated 938 stars on GitHub, and is licensed under the MIT License.
• Key Features:
  1. 13-step automated threat analysis pipeline: Enter keywords such as hacker groups or malware, and the system automatically searches and scrapes the Tor network, using LLMs to generate structured reports and entity relationship graphs.
  2. Four-layer mandatory content filtering mechanism: Automatically blocks harmful or sensitive content during query, URL validation, content scanning, and entity filtering stages to protect investigators' mental well-being.
  3. Flexible LLM integration: Supports Ollama (enabling fully offline, air-gapped deployment), as well as cloud APIs like Groq, OpenRouter, and Google AI Studio.
• Quick Start: Before launching with a single command using Docker Compose, you must first set up the environment variables:

  git clone https://github.com/KatrielMoses/voidaccess.git
  cd voidaccess
  cp .env.example .env
  # 編輯 .env 填入你的 GROQ_API_KEY 或設定本地 Ollama 伺服器端點
  docker compose up -d
  

  Once launched, you can access the complete service, which includes Postgres, a Tor proxy, a FastAPI backend, and a Next.js frontend.
• Comparison: Compared to expensive annual commercial dark web intelligence platforms like Recorded Future or DarkOwl, VoidAccess is completely free and open-source. Most importantly, it supports local self-hosting and air-gapped deployment, ensuring that investigation processes and sensitive keywords are never leaked.
• Caveats: Connection speeds on the Tor network are slow and highly unstable, often requiring node rotation. Additionally, analysis reports generated by LLMs still carry a risk of hallucination, so critical threat intelligence still requires secondary human verification.
• Who it's for: Security teams and independent threat analysts with limited budgets who need to monitor dark web leaks.
• Link: https://github.com/KatrielMoses/voidaccess

---

2. Fornjot: An Open-Source Parametric CAD Engine for 3D Modeling with Rust Code

• What it is: It solves the pain points of traditional CAD software, such as closed file formats, inability to use Git for version control, and difficulty in automated generation via code. It allows engineers to write Rust code directly to define 3D models, achieving a "code-as-model" development experience.
• Recent Buzz: Currently with around 4,800 stars on GitHub and licensed under AGPL-3.0, it is a highly active Rust project in the open-source CAD-as-code space.
• Key Features:
  1. Parametric feature modeling: Fully define geometries and constraints using Rust. Modifying a single variable automatically rebuilds the entire 3D model.
  2. Standard format export: Supports exporting models to industry-standard formats like STEP and STL, making it easy to import into 3D printing or traditional CAD software.
  3. Modern Rust ecosystem integration: Can be lightly embedded as a library into other Rust projects, allowing developers to build custom 3D generation tools.
• Quick Start: You can install and run the example directly via Cargo:

  cargo install fornjot
  # 或者複製專案並執行內建的 spacer 範例模型
  git clone https://github.com/hannobraun/fornjot.git
  cd fornjot
  cargo run -p fj-app -- --model models/spacer.rs
  

• Comparison: Compared to expensive and closed commercial subscription software like Autodesk Fusion 360 or SolidWorks, Fornjot is completely free and open-source. It allows 3D models to be seamlessly integrated into Git CI/CD workflows, which is unimaginable with traditional CAD tools.
• Caveats: The project is still in its early stages of development, with frequent API changes. It currently lacks the capability to model complex surfaces and organic shapes, meaning it cannot fully replace traditional GUI CAD software just yet.
• Who it's for: Engineers who prefer solving problems with code and need precise parametric modeling (such as 3D printed parts).
• Link: https://github.com/hannobraun/fornjot

---

3. Gyroflow: An Open-Source Video Stabilization Software Based on Gyroscope Data

• What it is: It solves the pain points of traditional video stabilization (such as Premiere's Warp Stabilizer) during video editing, which is extremely time-consuming because it analyzes screen pixels and is prone to image distortion in low light or during intense movement. It directly reads the gyroscope data built into cameras or drones to perform precise and fast physical-level stabilization.
• Recent Buzz: With over 6,700 stars on GitHub and licensed under GPL-3.0, it has become a standard tool in the open-source video editing and professional drone communities.
• Key Features:
  1. Precise gyroscope data calibration: Directly reads gyro data from GoPro, Sony, Insta360, or blackbox (Betaflight) logs, achieving more precise, distortion-free stabilization than pixel analysis.
  2. Cross-platform GPU acceleration: The core algorithm is implemented in Rust and supports hardware acceleration such as Vulkan, Metal, and Direct3D 11, providing smooth real-time previews and high-speed exports.
  3. Professional-grade fine-tuning parameters: Offers complex dynamic zooming, lens distortion correction, and independent stabilization weight adjustments for different axes (yaw, pitch, roll).
• Quick Start: macOS users can install it directly via Homebrew:

  brew install --cask gyroflow
  

  Windows and Linux users can download the pre-compiled, portable zip package from the GitHub Releases page, extract it, and run it directly.
• Comparison: Compared to the built-in video analysis stabilizers in Adobe Premiere Pro or DaVinci Resolve, Gyroflow is several times faster. Because it is based on physical sensor data, its stabilization quality in high-motion scenes far surpasses traditional algorithms.
• Caveats: Your recording equipment must support logging gyroscope data (or use an external logger), and syncing the video with the gyroscope timeline sometimes requires manual calibration. Beginners may need to spend some time getting used to its parameter logic.
• Who it's for: Drone enthusiasts, action camera users, and professional video creators who demand extremely high video stabilization quality.
• Link: https://github.com/gyroflow/gyroflow

Conclusion

As seen from the tools introduced today, the open-source community is constantly leveraging innovative technical architectures, such as Rust's high performance and AI automation, to break the monopoly of commercial software. Whether in cybersecurity defense, hardware design, or video creation, these tools demonstrate immense practical value. I'm Muyan, and I'll see you tomorrow in the next "Open Source Daily"!