Developer tools
Create and update Linear issues with your agents.
Agents file issues from bug reports and alerts, triage the backlog, move work between states, and answer questions in comments. Linear events can start workflows when issues, comments, project updates, cycles, documents, or customer requests change.
5,000 free credits to start.
Agent reach
Linear GraphQL API
The full surface, searched at run time.
Connect with
- API key
- Your own OAuth app
Give the API key the Admin permission, or turn on Admin access for an OAuth app, so Helios can register the webhook for Linear events.
Event triggers
Yes · 5 events
Plus manual, schedule, and signed webhook.
What the agent can do
What agents do with Linear
Helios searches the Linear GraphQL API at run time. The agent finds the endpoint the task needs and calls it.
- Query and update Linear through its GraphQL API, within your connection's permissions.
- Create issues with a team, labels, priority, assignee, and project.
- Move issues between workflow states and update priority, estimate, or cycle.
- Comment on issues, project updates, and documents, and reply to comments.
- Read cycles, projects, and roadmaps to report on progress.
- Start a run when an issue, comment, project, cycle, document, initiative, customer request, or release is created, updated, or removed, or when an issue's SLA is at risk or breached. Available events depend on your Linear plan and permissions.
Examples
Tasks you can hand to a Linear agent
Triage every new issue
The agent reads the report, finds related issues and recent changes, then sets labels, priority, and an owner.
On each new issue in the Platform team: label it, set priority, and assign an owner.
- linear·read the issue and team labels
- linear·search open issues for duplicates
- github·find recent commits in the affected code
- linear·set labels, priority, and assignee
PLAT-482 triaged
- Labeled `bug` and `billing` · priority High
- Linked to PLAT-455, which reports the same error
- Assigned to the author of the last change to the invoice job
Start work when an issue moves to In Progress
When an issue moves to In Progress, the agent gathers the context an engineer needs and posts it as a comment.
When an issue moves to In Progress, post the related code, issues, and docs as a comment.
- linear·read the issue and its history
- github·search the code it mentions
- notion·find the design doc
- linear·comment with a starting brief
Brief posted on PLAT-490
- Three files that handle webhook retries
- The design doc section on idempotency keys
- Two closed issues with the same root cause
Answer questions in issue comments
When someone asks a question in a comment, the agent looks it up and replies in the same thread.
When a comment asks the agent a question, answer it on the same issue.
- linear·read the comment and the issue
- github·check the related pull request
- linear·reply in the thread
Replied on PLAT-501
- The fix is in the pull request that merged yesterday
- It ships in the next release
Summarize the cycle
A scheduled agent reads the active cycle and tells the team what shipped, what slipped, and what is blocked.
Every Friday, summarize the active cycle and post it to Slack.
- linear·list issues in the active cycle
- code·group by state and owner
- slack·post the summary to #platform
Cycle 42 summary
- 18 of 24 issues done
- 4 moved to the next cycle
- 2 blocked on the billing migration
Event triggers
Start a run on Linear events
Attach any of these events to a workflow and filter on the fields that matter. Manual, schedule, and signed webhook triggers work too.
| Event | What it means |
|---|---|
issue.created | An issue is created in a team. |
issue.updated | An issue changes. Filter on the state it moved to, such as In Progress or Done. |
comment.created | Someone comments on an issue, project update, or document, so an agent can answer or pick up the follow-up. |
project_update.created | A project lead posts an update, so an agent can check it against the issues in the project. |
issue_sla.high_risk | An issue is close to breaching its SLA, so an agent can escalate it before it does. |
Works with
Linear works with the rest of your stack
One agent, one run, several services. It reads from one tool and writes to the next in the same sandbox.
FAQ
Questions, answered
Put an agent on Linear.
Connect it once and describe the task in chat. Then reuse the prompt as a workflow on a schedule or a webhook.