| name | jira-read-ticket |
| description | Fetch a Jira ticket via the Atlassian MCP and summarize its intent and acceptance criteria. Use when: reading a Jira ticket, pulling ticket context, looking up a ticket's details, getting acceptance criteria. Building block for other skills that need ticket context. |
| argument-hint | Jira ticket key (e.g. DT-1234), or text/branch/PR to extract one from |
Jira Read Ticket
Fetch a Jira ticket and produce a concise, structured summary of its intent. This is a building block — other skills call it whenever they need ticket context.
Step 1: Resolve the ticket key
Find a Jira key (e.g. DT-1234, PROJ-567 — typically uppercase letters, a hyphen, and digits) from, in order of preference:
- An explicit key passed as the argument
- The provided text (PR title/body, branch name, commit message, user message)
If no key is found, report that no ticket could be identified and stop — let the caller decide how to proceed (ask the user, fall back to other context, etc.). Do not invent or guess a key.
Step 2: Fetch the ticket
Retrieve the full issue with the getJiraIssue MCP tool. Capture:
- Summary (title) and current status
- Description
- Acceptance criteria
- Linked issues and any comments or attachments worth noting
If the fetch fails (no access, bad key), report that clearly rather than fabricating details.
Step 3: Summarize
Output a brief, structured summary the caller can use directly:
**<KEY>: <summary>** (<status>)
<2–4 sentence plain-English description of the goal>
**Acceptance criteria:**
- <criterion> (or "none stated")
**Notes:** <constraints, dependencies, linked tickets — omit if none>
Keep it tight. The point is shared, reliable ticket context, not a transcript of the ticket.