Skip to main content
Early Preview. Agent Tasks is in early development and not yet generally available. Capabilities and setup will change before release.

​
Overview

Picks up a Linear issue and proposes the corresponding code change as a pull request against the repository (or repositories) you connect.

​
Setting It Up

1

Create the agent task

From the Agent use cases section when creating a new service, pick Linear Coding Agent (see Creating an Agent).
2

Configure Linear access (optional)

The template ships with one environment variable for connecting to Linear: an API key. If you don’t need Linear access for your workflow, remove it and adapt the instructions accordingly.To create the API key:
  1. Click your workspace name in the top left of Linear, then Settings
  2. Go to Security & access
  3. Under Personal API keys, create a new key
3

Connect the repository

Add the Git repository, or repositories, this agent should code against as its Context. You can also add a Qovery service as Context, alongside its repository.
4

Choose a Provider

Enter your Anthropic API key, the only provider supported today.
5

Add MCP Servers (Optional)

Not required for this use case by default, add any if your workflow needs them. See Add MCP Servers.
6

Add a trigger

Select From a webhook, and copy the generated webhook URL from the agent task’s overview page, you’ll need it next.
7

Point Linear at the webhook

In Linear, go to Settings > API > Webhooks and create one: add a description, paste the webhook URL from the previous step, and under Data change events, select Issues.
Unlike Jira, Linear’s webhook can’t filter which change on an issue triggers it, every issue event is sent. Filter by adding a command to the agent’s instructions instead, for example: “If the ticket doesn’t have the label execute-agent, don’t start the work.”
8

Add an output (optional)

To have the agent post to Slack when it’s done, add a webhook output pointing to a Slack Incoming Webhook URL. See Outputs for the setup steps.
See Configuration for resources, governance, and environment variables shared by every agent task, and Execution Mode for choosing how it runs.