AgentThreatBench – Benchmark for AI Agent Memory Security
First OWASP-backed security layer for ASI06 memory poisoning in agentic AI.
Secure memory management for the paranoid.
Memory protection crate with guard pages and zeroization, like memsec but with secrecy-style ergonomics.
Rust developers building password managers, crypto wallets, or credential handlers
memsec · secrecy
I've been building a project that handles high-value credentials in-process, and I wanted something more robust than just zeroizing on drop. A comment on a recent Show HN[0] made me realize that awareness of lower-level memory protection techniques might not be as widespread as I thought.
The idea here is to pull out all the tools in one crate, with a relatively simple API. * mlock/VirtualLock to prevent sensitive memory from being swapped (eg the KeePass dump) * Core dump exclusion using MADV_DONTDUMP on Linux & Android * mprotect to minimize exposure over time * Guard pages to mitigate under/overflows
After some battle testing, the goal here is to provide a more secure memory foundation for things like password managers and cryptocurrency wallets.
This was a fun project, and I learned a lot - would love any feedback!
First OWASP-backed security layer for ASI06 memory poisoning in agentic AI.
Impressive Rust substrate layer, but crate collection without proof of integration or adoption.
28 deconvolution algorithms in Rust with proper PSF modeling and ndarray support.
Modular Rust crates let you embed the editor core while keeping VS Code extensions.
Serde-for-agents model: decouple traits from implementations, compose what you need without the monolith.
Secure Enclave vault exec beats .env files for AI agent secret injection.