In the wild
These articles, tutorials, and videos have been created by independent authors outside of nolabs. Clicking a card will take you to an external site — we are not responsible for its content.
Linux creator Linus Torvalds tells AI haters to walk away from Linux, or go fork it
Reporting on Linus Torvalds' declaration that Linux is not an anti-AI project, the piece turns to co-founder Luke Hinds for perspective on running AI coding agents safely.
Introducing GuardDog 3.0: A new rules engine, transparent sandboxing, and more
GuardDog 3.0 migrates from Semgrep to YARA for faster package malware detection and introduces built-in sandboxing via nono-py to isolate the scanning process itself against potential vulnerabilities.
How I Sandbox my AI Agents
A walkthrough of using nono as a lightweight alternative to Docker for kernel-level permission sandboxing across multiple AI coding agents — allowing unsupervised operation during development.
Multi-Layer Sandboxing of AI Workloads in Kubernetes
Explores securing AI agents in Kubernetes by combining VM isolation with seccomp and Linux Security Modules, building a multi-layered sandbox that protects both host infrastructure and workloads.
Sandboxing Claude Code CLI with nono
A practical guide to sandboxing the Claude Code CLI using nono's kernel-level enforcement, covering installation, configuration, and day-to-day usage.
nono: the ultimate coding agent security tool — sandbox and supercharge Claude Code in just two steps
Introduces nono as a security sandbox for Claude Code, replacing noisy permission prompts with kernel-level enforcement that makes unauthorised system access structurally impossible.
From 'oh no' to nono — building apps on OpenShift with nono and Claude Code
Demonstrates sandboxing Claude Code with nono on Red Hat OpenShift — covering Linux Landlock isolation, session auditing and rollback, and a worked example deploying a Flask app to a container platform.
Using Swival with nono
Integration guide showing the simplest way to run nono alongside Swival — covering sandbox mode, filesystem boundaries, network controls, rollback, and credential handling for LLM providers.
Always Further — nono
Coverage of nono and the always-further team's approach to kernel-enforced capability sandboxing for AI agents.