Dashboard
See all your AI agents at a glance. Status badges, diff stats, tags, and categories give you a complete picture without opening a single terminal.
All 17 features, grouped by category. Screenshots are shown where one exists; the rest are listed as text-only cards.
See all your AI agents at a glance. Status badges, diff stats, tags, and categories give you a complete picture without opening a single terminal.
Full xterm.js terminal per session with real-time output streaming. Attach and reprompt any agent without leaving the browser.
While watching an agent's browser in the Browser tab, click Inspect, then click any element in the live view.
Per-session git diff with VCS context. See exactly what each agent has changed before you approve or merge.
Real-time alerts when agents need attention — approvals pending, sessions paused, or errors detected — without polling.

Type to filter sessions in real time across titles, paths, branches, and tags. "payment" narrows 6 agents down to 2 immediately.

Filter by status to surface agents that need your attention — spot every Needs Approval gate without scrolling through the full list.

Switch between 8 grouping strategies — tag, category, status, branch, path, program, session type, or flat list. A single session can appear under multiple tags simultaneously.
Manage multiple project contexts from one UI. Each workspace gets its own isolated session state, configuration, and approval history.
Select multiple sessions and act on them together — pause, resume, delete, or tag in one operation.

Teach Stapler Squad what's always safe for your project. Add a pattern like ^git log once and never review it again.

Structured triage for every pending agent action. Review diffs and approve or reject before anything reaches your codebase.
100+ built-in rules block dangerous operations automatically. Read-only commands, safe git operations, and common dev tasks are pre-approved out of the box.
Visualise decision trends and classifier performance over time. See which rules fire most, how often agents are approved vs. blocked, and where your review time is going.
Run one prompt across 2–5 isolated worktree sessions, compare their diffs side by side, and merge the best one.
Searchable, filterable record of every agent action across all sessions. Filter by session, time range, action type, or free-text search to find exactly what happened and when.
Live-tail application logs with time range picker, multi-select filters, export, and density controls. No need to SSH in to debug an agent.
Inspect your current Stapler Squad configuration directly from the UI — no hunting for JSON files.

Every agent gets its own git worktree. No branch conflicts, no dirty state bleed between sessions. Merge when you're ready.