How distributed Claude Code is built on a tunnelling service
- Source
- nfltr.xyz
- Author
- abbiya
- Date
Context
NFLTR describes a way to coordinate AI coding work across machines from one session. It builds on Anthropic's Claude Code, a coding assistant that can edit files and run commands. A coordinating session, called the hub, sends a task brief to a worker launched on a joined machine. That worker uses the machine's tools and project , then returns progress and results. This lets the hub assign work where the required files or services already reside.
According to the architecture account, the hub and workers connect outward to a relay that forwards messages; the participating machines need no inbound ports for this connection. The relay runs neither the model nor the task code. Capabilities remain opt-in on each machine, including permission for workers to execute shell commands.
- context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
- multi-agent — Using several AI agents on one problem — splitting work in parallel, checking each other, or filling different roles like planner and reviewer.
- AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
Checking sign-in…
Loading comments…