Anyone who has kicked off a long task in Claude Code or OpenAI's Codex knows the feeling: you hit enter, the agent disappears into its work, and you sit there wondering if it is halfway done or stuck in a loop. ClawJetty is a new lightweight tool that solves exactly that problem.
The concept is dead simple. Before an AI agent starts working, it creates a "run" through ClawJetty and hands you a public link. As the agent works, it posts short progress updates to that link in real time. You get a live timeline of what the agent is doing - which files it is reading, what decisions it is making, where it is in the process - instead of staring at a blank terminal for ten minutes.
ClawJetty is deliberately minimal. There are no accounts to create, no onboarding flow, no dashboard to configure. You give your agent access to the ClawJetty API, and it starts posting updates. The tool already works with Claude Code, OpenClaw, and Codex, which covers most of the AI coding agent market right now.
This fills a real gap. As AI agents take on longer, multi-step tasks - refactoring entire modules, debugging across multiple files, running test suites - the lack of visibility becomes a genuine productivity problem. You cannot course-correct an agent you cannot see. ClawJetty does not try to control the agent or add guardrails. It just makes the work visible, which is often all you need to decide whether to let it keep going or intervene early.
For solo developers and small teams running AI agents on non-trivial tasks, this is a practical addition to the workflow. The no-setup approach means you can try it in about two minutes.