Two critical Linux kernel privilege escalation vulnerabilities dropped in the past two weeks. Both are deterministic, single-command root exploits. One of them was discovered with AI assistance. The other had
I know what good software engineering looks like. I’ve been doing this for decades. But as a solo developer, I sometimes skip the process and rely on gut instinct and
SSH drops mid-session and your Claude Code context is gone. cs (Claude Session Manager) keeps sessions alive as persistent JSONL files, viewable and reconnectable across machines.
A lightweight Emacs popup directory browser that replaces the useless default C-x C-d binding with something you can actually navigate and open files from, without the full Dired overhead.
A TypeScript rewrite of paredit.js that fixes the original's mutation bugs, removes all runtime dependencies, and adds full type safety throughout the S-expression structural editing library.
One VS Code extension that consolidates Emacs keybindings, Dired, and paredit support — forked from vscode-emacs-mcx with zero runtime npm dependencies and no supply-chain surface.
A full digital asset management system built with Next.js and Express. Handles images, social posts, and video reels across multiple platforms with a single interface. Running in production.
A Docker service that monitors IMAP mailboxes, transcribes voicemail audio attachments with OpenAI Whisper, and forwards you the email with the transcription at the top instead of an audio file.
Standard fail2ban bans just get waited out. saslfail uses escalating bans — 48 hours, 8 days, 32 days — backed by a persistent SQLite database that survives restarts and remembers
FSE template edits can land in the database or in your theme files — they look identical in the editor. Understanding the difference matters the moment you try to deploy,
