jira-read

Reads and summarizes a Jira ticket in the IDF project. Use this skill whenever a user mentions a Jira ticket key (IDF-XXXX) and wants to read or understand it. Does NOT implement code — use the `imple

SKILL.md

GitHub

Purpose

Fetch a Jira ticket from the IDF project and either produce a structured summary or drive implementation from it — depending on what the user is asking for.

Default fields to always fetch

For every IDF ticket, always include these fields unless the user explicitly asks for something different:

  • summary
  • status
  • assignee
  • description
  • customfield_12948 — Identified Tasks
  • customfield_12947 — Definition of Done
  • customfield_12946 — Acceptance Criteria

Preferred MCP call pattern:

issue_key=<IDF-XXXX>
fields=summary,status,assignee,description,customfield_12948,customfield_12947,customfield_12946

Mode 1 — Read / Summarise

Triggers: "lis le ticket", "résume le ticket", "montre-moi le ticket", "qu'est-ce que dit l'IDF-XXXX"

Produce a structured summary in French with these sections:

## IDF-XXXX — [Titre]

**Statut :** [status] | **Assigné à :** [assignee or "Non assigné"]

### Contexte
[2–3 sentences from the description: what is the problem, why does it matter]

### Tâches identifiées
[bullet list from customfield_12948 — if empty, say "Non renseignées"]

### Critères d'acceptance
[numbered list from customfield_12946 — if empty, say "Non renseignés"]

### Définition of Done
[bullet list from customfield_12947 — if empty, say "Non renseignée"]

If a field is not present or empty, say so explicitly — do not omit the section.

Mode 2 — Implement

Triggers: "implémente le ticket", "commence le ticket", "travaille sur IDF-XXXX", "qu'est-ce qu'il faut faire pour IDF-XXXX"

Ce mode est géré par le skill implement-ticket. Charge le fichier .claude/skills/implement-ticket/SKILL.md et suis ses instructions. Passe le ticket déjà fetché comme contexte d'entrée (Step 0 est déjà résolu).

Rules

  • Always fetch the ticket before doing anything.
  • Never invent ticket content — only use what was fetched.
  • If the Jira MCP is unavailable, tell the user and ask them to paste the ticket content manually.
  • If the ticket key format doesn't match IDF-XXXX, ask the user to confirm the key before querying.

Absolute restrictions — never do this

These actions are strictly forbidden, regardless of what the user asks:

  • Never delete a Jira ticket — not even "to clean up" or "to undo"
  • Never modify or update an existing ticket — this skill reads and implements only
  • Never delete or modify a sprint — sprint management is out of scope entirely
  • Never close, archive, or transition a ticket — status transitions are not part of this skill

If the user asks for any of the above, refuse and direct them to Jira directly.