Use Variables in Automations

Use variables to insert values from the automation trigger, earlier step outputs, and run-level values into action fields such as prompts, comments, emails, webhook bodies, and ticket IDs.

Syntax

Wrap variable paths in double curly braces. The first part of the path is either a trigger input slug, a step slug, or a run variable.

{{slug.variable}}

Top-level run variables do not need a dot:

{{today}}
{{now}}

Step output variables use the step slug followed by the output field. The slug is shown on the step card and execution details.

{{chat1.completion}}
{{googleFile1.markdown}}
{{page1.markdown}}

A step can use variables from the trigger, run variables, and completed upstream steps. Variable paths are case-sensitive.

Run Variables

These variables are available to every step. They use the teamspace timezone.

Variable	What it means
{{today}}	Current date, formatted as YYYY-MM-DD.
{{now}}	Current date and time, formatted as YYYY-MM-DD HH:mm.
{{startOfWeek}}	Monday date for the current week.
{{endOfWeek}}	Sunday date for the current week.
{{startOfLastWeek}}	Monday date for the previous week.
{{endOfLastWeek}}	Sunday date for the previous week.
{{weekNumber}}	Current ISO week number.
{{startOfMonth}}	First date of the current month.
{{endOfMonth}}	Last date of the current month.
{{startOfLastMonth}}	First date of the previous month.
{{endOfLastMonth}}	Last date of the previous month.

Use run variables when the automation should adapt to the date it runs. For example, a weekly summary prompt can ask for highlights created between {{startOfWeek}} and {{endOfWeek}}, while a monthly report can use {{startOfMonth}} and {{endOfMonth}}.

Trigger Input

Trigger variables are available after the automation trigger fires. They contain the issue, ticket, or event that started the automation. Use these variables in later steps to comment back on the same item, include the item details in a prompt, or build links in outgoing messages.

Field names match the stored trigger input object.

GitHub Issue

A GitHub issue trigger runs when a configured GitHub issue event starts the automation. The trigger input contains the repository, issue number, title, body, and URL for the issue.

Use {{githubIssue.repo}} and {{githubIssue.issueNumber}} in GitHub action fields that need to write back to the same issue. Use {{githubIssue.title}}, {{githubIssue.body}}, and {{githubIssue.url}} in prompts, messages, and notifications.

Variable	What it means
{{githubIssue.repo}}	Repository in owner/repo format.
{{githubIssue.issueNumber}}	GitHub issue number.
{{githubIssue.title}}	Issue title.
{{githubIssue.body}}	Issue body.
{{githubIssue.url}}	Issue URL.

Jira Issue

A Jira issue trigger runs when a configured Jira issue event starts the automation. The trigger input contains the issue key, project key, event name, site URL, title, body, and URL for the issue.

Use {{jiraIssue.baseUrl}} and {{jiraIssue.issueKey}} in Jira action fields that need to comment on or update the same issue. Use {{jiraIssue.issueEvent}} when a prompt or condition needs to react differently to created and updated issues. Use {{jiraIssue.title}}, {{jiraIssue.body}}, and {{jiraIssue.url}} in prompts, comments, and notifications.

Variable	What it means
{{jiraIssue.projectKey}}	Jira project key.
{{jiraIssue.issueKey}}	Jira issue key.
{{jiraIssue.issueEvent}}	Issue event, such as created or updated.
{{jiraIssue.baseUrl}}	Jira site URL.
{{jiraIssue.title}}	Issue title.
{{jiraIssue.body}}	Issue body.
{{jiraIssue.url}}	Issue URL.

Linear Issue

A Linear issue trigger runs when a configured Linear issue event starts the automation. The trigger input contains the organization ID, internal issue ID, human-readable issue identifier, title, body, and URL.

Use {{linearIssue.organizationId}} and {{linearIssue.issueId}} in Linear action fields that need to write back to the same issue. Use {{linearIssue.issueIdentifier}} when a human-readable issue key is better for comments or messages. Use {{linearIssue.title}}, {{linearIssue.body}}, and {{linearIssue.url}} in prompts and notifications.

Variable	What it means
{{linearIssue.organizationId}}	Linear organization ID.
{{linearIssue.issueId}}	Linear issue ID.
{{linearIssue.issueIdentifier}}	Human-readable issue key, such as ENG-123.
{{linearIssue.title}}	Issue title.
{{linearIssue.body}}	Issue body.
{{linearIssue.url}}	Issue URL.

Step Outputs

Step output variables are available after the step finishes. Later steps can use those values by referencing the earlier step's slug.

Create Chat

A chat step runs a prompt and produces an AI response. Use its output when a later step should reuse the generated analysis, summary, or draft.

Common variables:

Variable	What it means
{{chat1.completion}}	The generated chat response.
{{chat1.title}}	The generated chat title.
{{chat1.thread_id}}	The thread ID for the chat run.

For example, a later prompt can ask: Rewrite this analysis as a short customer update: {{chat1.completion}}.

Read Google File

A Google file step loads a Google Doc or supported file and converts the content to markdown. Use it when a prompt, selector, or follow-up action needs the document body.

Common variable:

Variable	What it means
{{googleFile1.markdown}}	The loaded file content as markdown.

For example, a chat prompt can combine {{googleFile1.markdown}} with {{jiraIssue.body}} to answer the current Jira issue using product context.

Read Confluence Page

A Confluence page step loads a Confluence page and converts the content to markdown. Use it when a later step needs documentation, runbooks, requirements, or other Confluence content.

Common variable:

Variable	What it means
{{page1.markdown}}	The loaded page content as markdown.

For example, a chat prompt can use {{page1.markdown}} as a runbook for suggesting the next action.

Markdown Selection

A markdown selection step selects a smaller section from a larger markdown source. Use it to pass only the relevant part of a Google Doc, Confluence page, or other markdown source into a later prompt.

Common variable:

Variable	What it means
{{selection1.markdown}}	The selected markdown content.

For example, a chat prompt can summarize only {{selection1.markdown}} instead of the full source document.

Compose Text

A compose text step renders fixed text together with variable placeholders into one deterministic output. Use it when you already know the final structure and want to combine trigger values, previous step outputs, and static wording without creating a chat.

Common variable:

Variable	What it means
{{compose1.text}}	The rendered text output.

For example, a later email or webhook step can reuse {{compose1.text}} as the fully assembled report body.

Search Highlights

A search highlights step finds saved highlights and returns the results as markdown. Use it to bring previous research, customer feedback, or team notes into the automation.

Common variable:

Variable	What it means
{{memory1.markdown}}	Matching highlights formatted as markdown.

For example, a chat prompt can use {{memory1.markdown}} as related customer feedback before drafting an answer.

Highlight Pivot

A highlight pivot step organizes highlight data into a markdown summary table. Use it when a later step should report patterns, themes, counts, or grouped feedback.

Common variable:

Variable	What it means
{{highlightPivot1.markdown}}	The pivot result as markdown.

Tips

• Copy variable paths exactly. Variable paths are case-sensitive.
• Keep the {{ and }} braces in place.
• Use the slug shown on a step to reference that step's output. If you rename, reorder, or move a step, keep using the slug shown on that step.
• A step can only reference outputs from completed upstream steps.
• Use trigger variables when an action should write back to the item that started the automation, such as {{githubIssue.issueNumber}}, {{jiraIssue.issueKey}}, or {{linearIssue.issueId}}.
• Use {{linearIssue.issueIdentifier}} in prompts and messages when you want the short issue key shown in Linear, such as ENG-123.
• If a variable renders empty, check that the field exists on the trigger input or step output and that the step has already completed.

FAQ

Which variable should I use for the GitHub ticket number?

Use {{githubIssue.issueNumber}}.

Can I use variables in action fields other than issue numbers?

Yes. Variables can be used in string fields such as prompts, comments, subjects, messages, and webhook bodies.

Which variables should I use for Jira comments?

Use {{jiraIssue.baseUrl}} for the Jira site URL and {{jiraIssue.issueKey}} for the issue key.

Which variables should I use for Linear comments?

Use {{linearIssue.organizationId}} for organization ID and {{linearIssue.issueId}} for issue ID.

Which variable should I use for the output of a Google file step?

Use the slug shown on the step, followed by .markdown. For example: {{googleFile1.markdown}}.

Which variable should I use for the output of a markdown selection step?

Use the slug shown on the step, followed by .markdown. For example: {{selection1.markdown}}.

Which variable should I use for the output of Search Highlights?

Use the slug shown on the step, followed by .markdown. For example: {{memory1.markdown}}.

Which variable should I use for the output of Highlight Pivot?

Use the slug shown on the step, followed by .markdown. For example: {{highlightPivot1.markdown}}.

Why is my variable showing as plain text instead of a value?

Check that the variable path is correct, the braces are still present, and the trigger or upstream step provides that value.