---
title: SLAs in Venom Tasks
description: Give a team's issues a deadline by priority and label, in hours, days, business hours or business days, counted from creation or from leaving Triage; see the time left on every list, board and issue, get told when one is at risk or breached, filter by it and chart breaches in Insights.
section: cloud
order: 84
audience: [user, operator, agent]
sources: [services/nodejs-cloud/lib/tasks/sla.ts, services/nodejs-cloud/lib/tasks/sla-pure.ts, services/nodejs-cloud/lib/tasks/sla-store.ts, services/nodejs-cloud/lib/business-hours.ts, services/nodejs-cloud/components/tasks/SlaChip.tsx, services/nodejs-cloud/components/tasks/planning/SlaSettings.tsx, "services/nodejs-cloud/app/tasks/api/teams/[id]/sla", services/nodejs-cloud/migrations/0244_tasks_sla.sql, services/nodejs-cloud/migrations/0245_tasks_team_sla.sql]
features: [cloud.tasks]
---

An SLA is a promise about how soon an issue gets done: an urgent bug within four business hours, anything labelled
`security` within two days. A team sets its rules once. From then on every matching issue shows how much time it
has left, turns amber when it is at risk and red when it is breached, and its assignee hears about both.

## Set the rules

Open **Settings → Teams → (your team) → SLAs** (`/tasks/settings/teams/{key}#sla`). Team admins and the
organisation's owners and admins can change them; everyone else sees them read only.

Each rule has:

| Field | What it does |
|---|---|
| Priorities | The priorities it covers (Urgent, High, Medium, Low, No priority). None picked: any priority |
| Label | Only issues with this label (a team label or a workspace label). Any label: no label needed |
| Within | How long, a whole number |
| Unit | **Hours** and **Days** count the clock. **Business hours** count only the team's business hours. **Business days** count whole working days: one business day is one daily window |
| Counted from | **Creation**, or **Leaving Triage**: the clock starts the first time the issue moves out of Triage, and an issue in Triage has no SLA yet. An issue that never was in Triage counts from creation |

Rules are checked in order and the first one an issue matches applies, so put the narrow rules (Urgent and
`security`) above the broad ones (any priority). A team has up to 50 rules. Save sends them all at once.

### Business hours

Turn on **Business hours** to set a time zone, the working days and a daily window (for example Monday to Friday,
09:00 to 17:00 in Europe/Berlin). The window follows the zone's local time all year, daylight saving included.

- **Business hours** without a window of your own count Monday to Friday, 09:00 to 17:00 UTC.
- **Business days** without a window count whole weekdays (Monday to Friday) in UTC.
- An issue created at night or on a weekend starts counting when the next window opens: four business hours from
  Friday 16:00 (with 09:00 to 17:00) is Monday 12:00.
- A window can end at midnight (`24:00`), and any days can be working days, Sunday included.

The same business-hours maths runs Venom Support's SLA policies.

### At risk

**At risk** is when an issue has that share of its time left: with 25% (the default), an eight-hour SLA turns at
risk after six hours. Choose **Never** to skip the warning.

## What changes it

Saving the rules recomputes every open issue of the team right away and says how many changed. An issue's SLA is
recomputed whenever something a rule reads changes: its priority, its labels, its status (leaving Triage, completing,
canceling) or its team. A canceled issue has no SLA.

A breach is final. Once an issue is breached, raising or lowering its priority, or changing the rules, doesn't undo
it, and Insights keeps counting it. An issue completed after its due time is breached at the due time.

## Where you see it

- **Lists and boards**: the **SLA** chip shows the time left ("3h left"), "At risk · 1h left" in amber or "SLA
  breached" in red. A met SLA shows nothing there. Turn the chip on or off under **Display → Properties → SLA**.
- **The issue view**: an **SLA** row in the properties, with the due time when you hover it.
- **Filter**: **Filter → SLA** with Breached, At risk, On track, Met and No SLA. "Is not" works too. A link to the
  list keeps it as `?sla=breached,atRisk`.
- **Insights**: the **SLA breaches** measure counts issues that missed their SLA, placed in the window by when they
  breached, sliced and segmented by team, assignee, priority or any other field ([Insights](/docs/cloud/tasks-insights.md)).

## Notifications

A sweep runs every minute. When an open issue passes its at-risk time, its assignee (or, with no assignee, the
member who created it) gets **is at risk of missing its SLA** in the Tasks inbox and the bell. When it passes its
due time, the sweep stamps the breach and sends **missed its SLA**. Each is sent once per due time; if a priority
change moves the times, the at-risk warning can come again for the new ones.

## API and agents

- `GET /tasks/api/teams/{id}/sla` reads a team's settings and `PUT` replaces them:

  ```json
  {
    "rules": [
      { "priorities": [1], "amount": 4, "unit": "businessHours", "start": "triaged" },
      { "labelIds": ["<security label id>"], "amount": 2, "unit": "days" },
      { "priorities": [2], "amount": 3, "unit": "businessDays" }
    ],
    "hours": { "tz": "Europe/Berlin", "days": [1, 2, 3, 4, 5], "start": "09:00", "end": "17:00" },
    "riskPercent": 25
  }
  ```

  The answer is `{teamId, sla, recomputed}`. A rule keeps its `id` when you send it back; a new rule gets one. A bad
  value is `400` with a sentence saying which, a member who may not change the team is `403`.
- Every issue carries `slaRuleId`, `slaStartedAt`, `slaRiskAt`, `slaDueAt` and `slaBreachedAt` (all null without an
  SLA).
- `GET /tasks/api/issues?sla=breached,atRisk` lists by SLA status at the time of the request; `sla=none` lists issues
  without one. In a filter tree it is the `sla` field ([Tasks API](/docs/cloud/tasks-api.md)).
- `GET /tasks/api/insights?measure=slaBreach` is the Insights measure.
- MCP: `tasks_sla_get` and `tasks_sla_set` read and replace the settings, `tasks_issue_search` takes `sla`, and
  `tasks_issue_get` shows an issue's SLA. Their REST twins are in the workspace API
  ([Workspace MCP](/docs/cloud/workspace-mcp.md)).
