At a glance
- Emits
- issues, count, hasMore
- Safe on a timer
- It only reads — no side effects
Set it up
Connect a Linear workspace in Settings, Integrations — see Integrations. Then choose a query, narrow it with the shared filters, and read {{issues}} downstream.
Configure
| Field | Default | What it does |
|---|---|---|
| Query | My assigned issues | My assigned issues, Current cycle, Project issues, or Search (filters only), which searches the whole workspace with nothing pinned. |
| Account | — | The connected workspace. Always shown. |
| Team | — | Whose cycle to read. Shown only for Current cycle. |
| Project | — | Which project to read. Shown only for Project issues — a project scopes on its own, with no team above it. |
| State | All states | Open is triage, backlog, todo and in progress; Active is todo and in progress. |
| Priority | Any priority | High or Urgent, or Urgent only. |
| Updated Within | Any time | Last 24 hours, 7 days or 30 days, on updatedAt. |
| Max Results | 50 | One page, clamped to Linear's 1–250 window. More matches than that sets hasMore. |
Output
| Key | Type | Value |
|---|---|---|
issues | Array | The matching issues, as real structure. Nothing matched is an empty array, never a missing key |
count | Number | How many came back on this page |
hasMore | Boolean | The workspace holds more matches than one page returned |
issues.0.identifier | String | Readable id of the first issue, e.g. WAT-123 |
issues.0.title | String | Its title |
issues.0.state.name | String | Its workflow state, e.g. In Progress |
issues.0.url | String | Its web URL |
Each issue carries identifier, title, url, priority (0 none, 1 urgent, 2 high, 3 medium, 4 low), priorityLabel, state.name, state.type, assignee.name, assignee.displayName, assignee.email, project.id, project.name, cycle.id, cycle.number, cycle.name, cycle.endsAt, team.key, team.name, dueDate, createdAt and updatedAt. There is no description — a briefing wants the shape, not the body.
Example: the morning briefing
A Schedule fires at eight, the query pulls your issues, and one AI Prompt reads the whole {{issues}} array and writes the paragraph a Notification delivers.
Paste this into the Flow Builder to get the same flow:
Every weekday at 8am, pull my open Linear issues and have an AI write a two-sentence stand-up summary grouped by state, then notify me with it.
Good to know
One AI call beats a loop for anything you will read — {{issues}} hands the model the real array. Reach for For Each with Loop Over set to issues only when each issue needs its own side effect; then read {{item.identifier}} and the rest.
Results land under issues, not linear, so a Create Issue earlier in the same flow never clobbers them.
An Agent can run the same query through the read-only linear_query_issues tool, with the same four modes and filters.