lab700xdev

6 ms·

1.
by lab700xdev 9mo ago
@altomek - Thanks for the suggestion! Just shipped v0.3.0 which includes a native GGUF header parser. It now extracts metadata and checks for license risks in .gguf files.
2.
by lab700xdev 9mo ago
This is incredibly valuable feedback. I’ve been reading through the pickle-fuzzer repo this morning, specifically about stack manipulation bypassing static heuristics. You nailed the trade-off: AIsbom is designed for the "90% hygiene&q
3.
by lab700xdev 9mo ago
That barrier to entry ("laziness") is the #1 security vulnerability. If it takes 3 minutes to set up a scanner, nobody does it. That's actually why we built the Web Viewer - so you can just drag-and-drop the JSON output rathe
4.
by lab700xdev 9mo ago
The comparison to npm is spot on. We are seeing the exact same pattern: a massive explosion of dependency complexity, but now the "dependencies" aren't 50KB JavaScript files, they are 10GB binary blobs that we treat as black
5.
by lab700xdev 9mo ago
Fair point on the terminology overlap with "Zip Bombs" (resource exhaustion). I used "Pickle Bomb" colloquially to describe a serialized payload waiting to detonate upon load, similar to how "Logic Bomb" is use
6.
by lab700xdev 9mo ago
You are right that the inference ecosystem (llama.cpp, vLLM) has moved aggressively to GGUF and Safetensors. If you are just consuming optimized models, you are safer. However, I see two reasons why the risk persists: 1) The Supply Chain Ta
7.
by lab700xdev 9mo ago
Thanks for the link! fickling is excellent work (and definitely the gold standard for deep analysis). The goal with AIsbom was to build something lightweight enough to run in a fast CI/CD loop that creates a standard inventory (Cyclone
8.
by lab700xdev 9mo ago
The Golden Rule holds: "Don't unpickle untrusted data." The problem I'm trying to solve is that "Untrusted" has become blurry in the AI age. Data Scientists treat Model Hubs (like Hugging Face) as trusted repos
9.
by lab700xdev 9mo ago
That is entirely fair feedback regarding the new accounts. We all have to start somewhere! That is exactly why I open-sourced the engine (Apache 2.0) and kept the logic in Python rather than a compiled binary - so you don't have to tru
10.
by lab700xdev 9mo ago
Ideally, CI/CD Pipeline (Pre-Merge) - We recently released a GitHub Action for this exact workflow. The goal is to block a Pull Request if a developer tries to merge a .pt file that contains CRITICAL risk opcodes. If you wait until Run
11.
by lab700xdev 9mo ago
You are absolutely right - blocklisting is a game of whack-a-mole. However, in the context of serialized ML weights, the "allowlist" of valid imports is actually quite small (mostly torch.nn, collections, numpy). Right now, we are
12.
AIsbom – open-source CLI to detect "Pickle Bombs" in PyTorch models(github.com)
52 points by lab700xdev 9mo ago|38 comments
13.
by lab700xdev 9mo ago
Hi HN, I’ve been working with ML infrastructure for a while and realized there’s a gap in the security posture: we scan our requirements.txt for vulnerabilities, but blindly trust the 5GB binary model files (.pt) we download from Hugging Fa