Skip to main content

/jira

Fetch a Jira issue and investigate it using the /questions workflow.

Usage​

/jira <Jira URL or issue key>

Accepts either a full Jira URL (e.g. https://turquoisehealth.atlassian.net/browse/CLD-7) or a bare issue key (e.g. CLD-7).

Prerequisites​

Add your Jira credentials to .env at the project root:

JIRA_EMAIL=your_email@turquoisehealth.com
JIRA_API_TOKEN=your_token

API tokens can be created at https://id.atlassian.com/manage-profile/security/api-tokens.

What It Does​

  1. Fetches the Jira issue — retrieves the issue summary, description, status, priority, assignee, reporter, comments, linked issues, and attachments via the Jira REST API. Text-based attachments (.sql, .csv, .md, .txt) are downloaded and read.

  2. Determines the domain — reads the issue content to infer which domain(s) apply (cld, spines, or consumer) and asks you to confirm. Does not assume based on the Jira project key alone, since tickets may cross domain boundaries.

  3. Runs the /questions investigation — formulates the issue summary, description, and comments into a question and follows the full /questions investigation flow for the confirmed domain. Since this was triggered from Jira, the questions skill will automatically post the answer back as a comment on the Jira issue (including any PR links if code changes were made).

  4. Saves the output — writes the investigation to .claude/skills/questions/answers/{date}_{issue_key}_{short_description}.md.

Key Files Read​

  • .env — Jira credentials (JIRA_EMAIL, JIRA_API_TOKEN)
  • All files used by /questions (see /questions documentation)
On this page: