Shuttle

Kalindu De CostaCode & Reviews
Get it for macOS

Puts every pull request that needs you, from any server, in one place. The Inbox lists the decisions waiting on you, your PRs keep the agent thread that fixes their reviews and failed checks, and the review centre groups PRs that ask you or your teams by where your review stands. Merge, comment and post AI review drafts from Shuttle, on GitHub out of the box or on any other server you describe with a command adapter. Insights show where review time goes: who reviews your PRs, how fast, and what your agents fixed.

Overview

View source

Shuttle puts every pull request that needs you in one place, whatever server hosts it: GitHub out of the box, and Delta, GitLab or any other server you describe with a few shell commands that print JSON.

Four places

  • Inbox: only the decisions waiting on you. Each card says why it is there, has one main button, and says when it leaves: your PR is approved and green, feedback on your PR has nobody on it, an agent asked you something, an AI review draft is ready to post, a PR changed after your review, or a new request asks for you by name or mentions you. Snooze hides a card until tomorrow, or until something new happens.
  • My PRs: your open PRs grouped by what happens next, each showing its checks, its reviews and its agent; then your merged and closed ones.
  • Reviews: other people's PRs that ask you or your teams, mention you, or that you track, grouped by where your review stands. Teams read the same on every server, and bots and PRs quiet for a month are hidden unless you show them.
  • Insights: how long your PRs wait and take to merge, how fast you review, what agents did with feedback, how many AI findings you post, and ideas that come with the button that acts on them. The automation log lives here too.

Agents

  • A PR's agent is one thread that gets its new reviews, comments and failed checks: Fix it asks the thread to fix them, Just tell it only sends them, Off stops. Give a PR an agent, change it, or start a new thread for it; the earlier one stays linked for reference.
  • Auto agent (off by default) makes the bb thread that opened your PR its agent, found by the thread id in the branch name or by its branch. Otherwise Shuttle suggests it.
  • Agents report what they did with each item, so the conversation shows "Fixed in 7c41d2e" or "Agent replied" beside each comment.

Acting from Shuttle

  • Merge your PRs (squash, merge or rebase, as the repository allows), only at the head you saw, or turn on auto-merge. Comment, reply in review threads, resolve them, approve or request changes, as yourself.
  • Start an AI review with a preset. The review thread hands its findings back as a draft: tick the ones to post, edit their text, pick a verdict and post one review with inline comments.
  • Track any PR by pasting its link, whether or not you are part of it.

Settings live on Shuttle's page in bb's Settings: servers (with a Test button), review presets, automations and agents, the sidebar section review threads go to, auto agent, whether new requests come to the Inbox, bot authors, and import or export as TOML. Agents get the shuttle_link, shuttle_unlink, shuttle_prs, shuttle_review_draft and shuttle_feedback_done tools, and bb shuttle covers the same from the command line.

More from Kalindu De Costa

BurnSee what every agent thread costs in dollars while it runs, per prompt and across all your threads. A chip beside the context circle shows the thread's total and what the last prompt added; click it for where the money went, a chart of every model call and the total with child threads. The Burn tab in the side panel lists the cost of each prompt, and a Burn page in bb's navigation charts spend across threads by day, project, model and harness. Budgets notify you at 80% and 100% of a thread's budget, and stop a run only if you turn that on. Works with pi, OpenCode, Codex, Claude Code and any harness that reports usage to bb; where a harness records no cost, Burn estimates it from list prices and marks it with ~.57LoomHand a goal to agents and keep the decisions. A planner turns it into a spec with tickets, acceptance criteria, checks and a merge plan, and a separate agent reviews the spec before you approve it. Workers then build each ticket in its own git worktree, Loom runs your checks, and reviewers on another model approve the work or send findings back. Approved work merges into an integration branch you land as one pull request, or opens as one pull request per ticket that Loom follows through CI. A full-screen Loom page shows what needs you, the board, every attempt and review, the changes before they land, insights on how the project went and a timeline you can rewind.NewSession LogCheck what an agent actually ran, changed and decided, in one live log beside the chat. Messages, thinking, commands with their output and exit codes, tool calls, file diffs and subagent steps appear in timeline order, with a timestamp on every row and a duration on every command. The log follows new output as it arrives, opens failed commands by default, and filters to messages, thinking, tools or errors. Search covers the whole session, and single keys jump between rows, replies and errors. Open it with Mod+Shift+L or from the thread header; it reads bb's timeline and keeps no copy of the thread.NewSwitchboardSee which agent threads failed, need you or are still running, grouped under your bb sections. Child threads nest under their parent across projects, and a collapsed parent still shows its most urgent child. A status-lanes view lists every awake thread under Failed, Needs you, Running, Unread or Idle. Drag threads to reorder, nest or move them between sections, and snooze a quiet family until tomorrow morning. Hover a row for the thread's model, context use, branch, uncommitted changes and pull request. Choose Switchboard under Settings → Appearance → Sidebar in each bb client.New

More in Code & Reviews 45

Work with repositories, builds, changes, pull requests, issues, and reviews.

View all