Features

A subscription isn't just "pipe events to a channel." It's a small, powerful control panel for exactly what gets delivered, where, and how loudly. Here's the full toolkit — free features and PRO ones alike.

Destinations

Each subscription delivers to exactly one place:

• Teams channel (via the bot) — team-wide visibility, supports in-place updates and incident collapsing.
• Incoming webhook — paste a Teams Workflows / Power Automate URL. No bot needed.
• Direct message PRO — the bot DMs a specific person (see On-call DMs).

Routing PRO

Send different events to different places from one subscription. The first matching route wins; if nothing matches, the default destination is used. Match on any combination of:

• pipeline — exact or * glob (e.g. deploy-*)
• branch — exact or * glob
• environment — exact (e.g. prod)
• status — any-of (e.g. ["failure", "cancelled"])

Classic pattern: prod failures to #prod-alerts, everything else stays in #ci.

Incident mode PRO

When a pipeline flaps, incident mode collapses its repeated failures into one card that updates in place — the first failure opens it, repeats bump a counter, and the next success resolves it. Instead of 20 red cards, you get one that tells the story. Channel destinations only.

On-call DMs PRO

The bot opens a 1:1 chat and DMs a specific person when their pipeline breaks — perfect for on-call ownership. The recipient adds the app once (opens a 1:1 with the bot), then appears in the destination picker. Pair it with routing to DM the owner only on prod failures.

Noise controls PRO

• Quiet hours — suppress deliveries during a recurring window (e.g. 22:00–07:00 in your timezone). Overnight windows wrap past midnight.
• Dedupe window — swallow a repeat of the same (pipeline, status) within N seconds. Tames a job that fires the same failure on a loop.
• Failure-only — deliver only failures plus the first success after a failure (the recovery). Kills steady-state green noise.

@mentions on failure PRO

Map owners to a subscription and the failure card @mentions them, so the right people get pinged in a busy channel. Identify people by Teams MRI or AAD object id.

Status filters

Only care about failures? Filter a subscription to just the statuses you want — failure, success, started, deployments, and more. Available on every plan.

Approvals PRO

Gate a deploy behind an approve/reject Adaptive Card right in Teams. The decision is recorded in the audit log with who clicked and when — a lightweight change-control step without leaving chat.

Custom branding PRO

Show your own logo on every card a subscription sends, in place of the default status glyph.

Reliability & audit

• In-place updates — on bot channels, a run's card updates from started → success/failure instead of stacking new cards.
• Retries — transient delivery failures are retried with backoff; timeouts are handled so you don't get duplicate posts.
• Per-org rate limits — a noisy org can't drown out others.
• Audit log — subscription changes and approval decisions are recorded for forensics.