Claude Code (subscription) has Agent Teams built in. Teams of Agents communicate with local files that they use as inboxes and task list. Has tmux and iTerm 2 integration.
https://code.claude.com/docs/en/agent-teams
They can rack up some extra tokens if you leave agents going idle. Because they loop, checking for new messages for them.
Obsidian shows a warning about this. But the only issue it's pointing out is that mixing Obsidian's built-in sync with something that syncs your files is likely to cause problems. Otherwise it's a perfectly safe and normal way to sync.
Using Claude Channels can make it easy to inject prompts and get just the response back without having to identify it in the terminal output or fight with the TUI.
But they're not well designed, and some things just have to go in through the terminal interface like slash commands (i.e. `/clear`)
The law of Software Envelopment. Jamie Zawinski in 1995 stated that "software inevitably expands to include email functionality, or it is replaced by software that does"
Clearly, that's not the case anymore. Nowadays you just swap out "email" for "LLM"
> As software engineers, I get the feeling we’re moving almost entirely away from code.
Wille-Driven Development — I prefer the word from Schopenhauer's Die Welt als Wille und Vorstellung (The World as Will and Representation).
What we will begins to matter more. The gap between desire and existence is collapsing. In Genesis, creation is a pure speech-act: fiat lux, and it was so — no implementation details, no build step, will and result simultaneous. We're not there yet. But I find the distance is measurable and shrinking.
The Representation — code, syntax, toolchains — is receding. The Will asserts itself more directly. Welcome to the future.
reply