SPERIXLABS
BUILD WITH SECURITY
SPERIXLABS is a cybersecurity research lab where academic research meets hands-on security practice. We publish peer-reviewed work on IoT security, 5G networks, blockchain, and AI security — and we break things carefully: mobile apps, routers, LLM agents, and more — then write up what we learn.
From the blog
- The Function That Beat the Model: What We Measured When We Removed the LLMs Sep 21, 2026 For months, a 1-billion-parameter model validated sensitive-data detections in our proxy pipeline. Last week we ran it against a 40-line …
- Manifold: Chaining Seven Proxy Layers Into a Single Transparent LLM Pipeline Jul 29, 2026 Every LLM proxy tool you install — a redactor, a router, a rate limiter, a cache — spins its own HTTP server. Each one speaks the OpenAI or …
- DAEDALUS — A Security Research Control Plane for Apple's Container Runtime Jul 28, 2026 What is DAEDALUS? DAEDALUS is an open-source security-research control plane built on Apple’s container runtime. It exploits a unique …
- Fake recruitment, remote JSON payloads, and function-inliner malware (analysis) May 19, 2026 Clone-and-run recruitment lures that pull obfuscated Node.js from jsonkeeper-style hosts: IOCs, SHA-256 hashes, a small fetch script, function-inliner teardown, MITRE mappings with links, and what to do if you already ran the artifact.
- LLM-Redactor: What Leaves Your Prompt When You Talk to a Cloud LLM Apr 15, 2026 Every time a coding agent sends a prompt to a cloud LLM, the full content of that prompt — your code, your credentials, your customer names, …
- Local-Splitter: Cutting Cloud LLM Costs by Putting a Small Model in Front Apr 15, 2026 Cloud LLM tokens are expensive. Not in the “my AWS bill is high” sense — in the “I’m burning $0.015 per 1K output tokens and my coding agent …