Most ADEs bring the browser into the IDE. tabsh takes the opposite approach: bring coding sessions to your browser, where your work already happens. It's a small Rust daemon that puts all your sessions across projects on one Kanban board/list, with status updates via agent hooks. Use real terminals or chat panels with Claude Code, Codex, Cursor CLI, OpenCode, and more. Free, MIT-licensed, macOS/Linux.
Maker here! My ADE was the last thing I kept ⌘-Tab-ing to. Issues, PRs, docs, design, and localhost were all in the browser. Why not bring the ADE there too?
tabsh is a small Rust daemon that brings all your coding-agent sessions into one browser board, across projects. Hooks update statuses automatically, including when an agent needs your input.
Everything runs locally. No account, no server, macOS and Linux.
Free, MIT-licensed. I'd love you to give it a try and fork it and make it your own.
The cross-project board is the useful inversion here. One status I'd make first-class is "stopped vs verified": a session can exit cleanly while its tests never ran, or pass and then have the tree change. If hooks can carry the verifier command, exit code, and final tree state into the card, the board becomes a review queue instead of just a session monitor.
I like the inversion here: instead of dragging the browser into the IDE, you bring the sessions to the browser where issues, PRs, and docs already live. A single Kanban view across projects sounds like the real win when several agents are running at once. How well do the hook-based status updates hold up when a session stalls or waits on a question, is that state obvious at a glance?
how does the daemon guards localhost against DNS rebinding and cross-origin requests ? does the token check also validates Origin and Host headers, or what's planned next ?
Nice one! The best thing about this is I have everything I need in one window, and context switching has become so much more manageable
Across several projects, I'd want sessions waiting on me to be the first thing I see. Hook-based status updates seem a useful way to keep that view current.
About tabsh on Product Hunt
“Work with agents with an IDE in your browser tab”
tabsh launched on Product Hunt on October 26th, 2026 and earned 102 upvotes and 10 comments, placing #12 on the daily leaderboard. Most ADEs bring the browser into the IDE. tabsh takes the opposite approach: bring coding sessions to your browser, where your work already happens. It's a small Rust daemon that puts all your sessions across projects on one Kanban board/list, with status updates via agent hooks. Use real terminals or chat panels with Claude Code, Codex, Cursor CLI, OpenCode, and more. Free, MIT-licensed, macOS/Linux.
tabsh was featured in Open Source (69k followers), Developer Tools (521.9k followers) and GitHub (41.5k followers) on Product Hunt. Together, these topics include over 140.3k products, making this a competitive space to launch in.
Who hunted tabsh?
tabsh was hunted by Blaise Tiong. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how tabsh stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Blaise Tiong
Maker here! My ADE was the last thing I kept ⌘-Tab-ing to. Issues, PRs, docs, design, and localhost were all in the browser. Why not bring the ADE there too?
tabsh is a small Rust daemon that brings all your coding-agent sessions into one browser board, across projects. Hooks update statuses automatically, including when an agent needs your input.
Everything runs locally. No account, no server, macOS and Linux.
Free, MIT-licensed. I'd love you to give it a try and fork it and make it your own.