syumei

I've heard of Vulnetic, and it looks quite cool, though I feel it's a bit expensive for individual users (probably fine for companies)

Oh..... sorry. the correct link is https://medium.com/@gosu0x/how-tiny-rust-layout-choices-free...

not yet (I'm in US)

I guess the next model's name might be "galaxy"

Hey, thanks a lot! I’d never really read the terms and conditions seriously, so I learned a lot!

Thank you so much for the kind feedback. Your point makes a lot of sense, and I apologize for focusing too heavily on my own projects recently. I love the HN community, and I will make sure to contribute more thoughtfully and meaningfully going forward.

The repository is available at https://github.com/h5i-dev/h5i/blob/main/crates/h5i-wasm-har...

That’s a fair point. The agent-facing functionality does overlap quite a bit with Jupyter-MCP-Server.

The main difference is that this also provides a terminal-native notebook UI for humans. For example, over SSH, you can inspect, edit, and execute cells, and even view plots directly in the terminal witout opening JupyterLab or setting up port forwarding. Agents can also operate on the same notebook through a regular CLI rather than requiring MCP.

Thank you so much!

Claude Code and Codex both run untrusted commands. This post compares how their sandboxes protect your repo, secrets, and host machine.

quite interesting post

Fair criticism. I don’t think the value is just "agents can send text to each other”; that part has many implementaton and design choices. The value of h5i I’m exploring is making the intermediate state reviewable: review requests, risks, handoffs, unresolved claims, associated prompts and AI-to-Ai conveersation, and final decisions tied to the branch/PR.

I've been tracking real-world AI agent failures and incidents recently.

Things like:

prompt injection → goal hijacking

agents going rogue due to misalignment

unintended/unsafe tool use

It feels like we're starting to see repeatable patterns, not just isolated bugs.

I'm collecting cases + papers here:

https://github.com/h5i-dev/awesome-ai-agent-incidents

If you've seen interesting incidents, weird failures, or relevant research, I would love to add them.

Hi I'm currently implementing a symbolic execution engine for EVM in Rust. Though my current implementation is just a translation of hevm, I would like to add some novel features. For example, I'm considering using A* search or multi-armed bandit to enhance scalability. Any kind of feedback, ideas, and requests are super welcome!!!

AIJack is an easy-to-use open-source simulation tool for testing the security of your AI system against hijackers. It provides advanced security techniques like Differential Privacy, Homomorphic Encryption, K-anonymity and Federated Learning to guarantee protection for your AI. With AIJack, you can test and simulate defenses against various attacks such as Poisoning, Model Inversion, Backdoor, and Free-Rider. We support more than 30 state-of-the-art methods. For more information, check our documentation and start securing your AI today with AIJack.

Hello, Hacker News community! I am excited to introduce a new project called Gymbo, a Proof of Concept for a Gradient-based Symbolic Execution Engine. Gymbo is designed to push the boundaries of symbolic execution by leveraging recent advancements in gradient descent to tackle SMT-like formulas. This approach allows Gymbo to discover input values that satisfy each path constraint during symbolic execution.

Gymbo is entirely implemented in C++ and relies only on standard libraries.

What sets Gymbo apart from other symbolic execution tools is its simplicity and compactness in implementation. I believe that this project will help individuals better understand the core principles of symbolic execution and SMT problem-solving through gradient descent.