Bands

A live plan Claude posts and keeps current: progress above the prompt, the full checklist and its history in a pane.

Search this project's past sessions: /recall for you (attach an old exchange to your next prompt), a tool for Claude to look up earlier decisions.

Works through a Markdown checklist item by item: one prompt per item, an optional check command, the box ticked when it passes.

Runs your dev server or watcher for the session, flags its errors above the prompt, and lets Claude read its output.

CI that fixes itself with your say-so: watches the branch's workflow run after a push, puts a failure with its error-dense log on the band with Fix it, surfaces new review comments, and can pre-review PRs you are asked to review.

The first session of your day starts with a briefing: what changed since you left, what waits on you, your work in flight, a first step and a standup draft.

Audits the project's dependencies once a day in the background (npm, pnpm, yarn, pip-audit, cargo audit) and flags high-severity vulnerabilities above the prompt, with details and a one-press safe upgrade by Claude.

Catches a broken dev environment before Claude trips on it: wrong Node/Python/Go versions, missing node_modules, .env gaps, merges or rebases in progress.

The things Claude told you to do yourself, never lost: captured from answers, kept per project, mirrored to .claude/follow-ups.md, one press to finish, snooze or hand back to Claude.

Notices when a session is getting heavy (full context, repeated compactions, slow turns, tool errors), prepares a handoff in the background, and offers a one-press fresh start that continues from it.

Claude gets better at your project without you writing rules: your corrections and the command errors Claude fixes become one-press lessons for CLAUDE.md.

Works your backlog while you are away: when idle, one worker at a time implements the next .claude/backlog.md task in its own git worktree under a strict tool policy and budget; in the morning you review, adopt or discard each branch.

Claude knows what you changed between turns: files edited outside Claude are named (with small diffs) beside your next prompt, and an optional cheap review flags slips in your own edits.

Runs the tests related to each change (Claude's edits and yours) in the background; a failure caused by Claude's edit reaches Claude mid-turn, and the band offers Show and Fix it.

Say "this error", "the tests", "this file", "what you just did" or "the plan" and what it points at goes along with your prompt: the failing output, the test failures, the file, the diff, the plan. No model calls, size-capped, never twice, and Remove takes it back.

Keeps work flowing: when Claude stops mid-task ("Next I'll…") or asks an obvious yes ("Shall I continue?"), a 5-second countdown you can stop sends a short continue. Never for choices, missing info or risky steps.

Every prompt you type gets a quiet second read: without changing your words, a hidden note gives Claude the goal in one line, the likely scope, the project's checks to keep green, and what done means. Haiku only for vague prompts; See, Undo and Off in the band.

A quick second pair of eyes after edits: reviews the turn's diff in the background and offers to fix what it finds.

A prompt queue that runs itself: line up prompts with /queue and they run one after another, pausing when Claude asks you something or a turn fails.

Sharper retries after a failed turn: when checks still fail, Claude gives up, errors loop, or you say "that didn't work", haiku drafts a retry with the lesson, a different approach and the exact failing output. One click sends it.

No lazy placeholders: flags TODOs, '... rest of code', NotImplemented stubs and empty bodies an edit introduces, nudges Claude to finish them, and lists what is left.

One prompt, several jobs: when a prompt you type holds independent tasks (a list, numbered steps, "do A, then B, and also C"), offers to run them one by one as their own turns, stopping at the first that fails or asks you something.

Runs your project's checks after Claude edits, shows pass/fail above the prompt, and closes the loop with one-press (or automatic, bounded) fixes.

Jev (TypeSafe's System One model) as Claude's instant co-processor: bulk filter/classify/rank/semantic-grep tools, risk notes under permission dialogs, a check on unverified 'it passes' claims, and optional per-prompt effort.