Writing

Notes, essays, and things worth writing down.

• 06 / 2026

  AI tooling is company infrastructure. Most teams still run it like personal dotfiles.

  Every employee runs their own local Claude. Anthropic shipped the shared install; the plugin repo to put through it is the part most teams skip.

  • #ai-tools
  • #plugins
  • #infrastructure
• 06 / 2026

  Your AI reviewer shares your AI coder's blind spots

  A Claude review proposed a bug. A second Claude review would have approved it. Same model family, same blind spots, so I review with a different one.

  • #ai-tools
  • #code-review
  • #verification
• 05 / 2026

  AI gives you consistent UI only after you build the system it copies from

  AI doesn't drift because it's careless. Each session starts blank. A component library, a CI drift gate, and a catalog of blessed exceptions hold the line.

  • #ai-tools
  • #design-systems
  • #enforcement
• 05 / 2026

  Hand every AI you use the same name tag

  Every new chat, you re-introduce yourself to your AI. Four plain files in ~/.me/ that any tool reads. One write, many readers, no vendor lock-in.

  • #ai-tools
  • #claude-code
  • #personal-context
• 05 / 2026

  Stop parsing TUI screenshots. claude agents --json is the HUD primitive.

  We hand-rolled a tmux status bar that watches Claude Code sessions by scraping terminal text. The 2.1.145 release shipped the JSON output we'd been faking.

  • #ai-tools
  • #claude-code
  • #observability
• 05 / 2026

  Cap your AI assistant's retry storms before the gate becomes the problem

  We wrote shell-script gates to enforce AI workflow discipline. Then the gates themselves started looping. CC 2.1.147 finally added the kill switch.

  • #ai-tools
  • #claude-code
  • #enforcement