Loop Desk is an always-on business workspace that takes incoming signal (email, forms, feeds, notes) and turns it into briefs, watch items, and next actions.

Does Loop Desk send emails on my behalf?

Loop Desk can draft responses, but nothing sends until you approve and explicitly trigger it. The approval boundary is fundamental.

What signal sources does Loop Desk support?

Email, RSS/Atom feeds, form submissions, manual notes, and webhooks. The desk treats all sources equally in its processing loop.

rev 207 — Loop Desk changelog

rev 2072026-08-11Accelerating

Makes the managed approval handoff measurable: immutable review clocks, 7/30-day cycle-time analytics, exact oldest-ready coverage, trend visualization, and typed API parity.

Immutable ready and decision clocks fix false queue freshnessNew additive artifact.readyAt and artifact.decidedAt timestamps define the human-review stage explicitly. Editing, tagging, sharing, pinning, or reacting no longer resets apparent review age. Every starter, AI-output, brief, ideation, single/bulk decision, status reset, and regenerate path stamps the lifecycle; historical rows fall back visibly to created/updated timestamps.
7/30-day ready-to-decision analyticsThe approvals panel now shows 7-day and 30-day median cycle time, 30-day p90, 72-hour breach share, 24-hour completion share, approved/archive outcome mix, and week-over-week direction. A 14-day daily-median bar chart makes review drift visible before it turns into a stalled queue.
Exact oldest-ready coverageThe dashboard preserves its newest-artifact slice but independently loads the exact oldest 30 ready outputs and appends any missing rows. A burst of unrelated briefs, edits, or share activity can no longer push a starving review out of the bounded dashboard result.
Throughput API and OpenAPI 3.1 parityGET /api/v1/artifacts now projects readyAt and decidedAt plus a workspace-wide approvalThroughput aggregate independent of list filters and limit. The contract includes both windows, percentile/share metrics, exact-vs-inferred sample quality, week-over-week trend, and stable 14-day buckets, extending typed coverage to its 115th unbroken rev.