sync-jiralisted

# Sync Jira ## When to use this skill Invoke when the user types `/tcgflow-sync-jira` or says "sync Jira", "refresh Jira status", "check the Jira status of our tasks", or when the Cockpit shows a stale "synced Xh ago". Also reasonable to run at the start of a work session or on a schedule (e.g. each morning). **Why this is an AI-run skill, not a Cockpit feature:** the Cockpit server is a zero-credential local file projection — it can't call Jira. Only the AI tool has the Atlassian MCP. So the AI fetches the statuses and writes a cache file; the Cockpit reads the cache. (ADR 0029.) ## Instructions You produce/update **one project-local file**: `.tcgstackflow/tasks/jira-cache.json`. Read-only on Jira. ### Procedure 1. **Collect Jira-keyed task IDs.** Scan `tasks/active/`, `tasks/completed/`, and `tasks/archive/` for task folders whose ID matches a Jira key pattern (`[A-Z][A-Z0-9]+-\d+`, e.g. `ES-6965`). Non-Jira IDs (e.g. `MIGRATE-GSF`, `BUG-flaky`) are skipped — they have no Jira ticket. 2. **Confirm the Atlassian MCP is available.** Check `claude mcp list` (or the tool's MCP list); `atlassian` is in `config.yaml`'s `mcp.recommended`. If it isn't connected, **stop and ask the user to connect it** — do not invent statuses. (Same discipline as the planner's Jira-fetch rule.) 3. **Fetch statuses.** Prefer a single JQL batch over N calls: `key in (ES-6965, ES-6900, …)` via the Atlassian MCP search, requesting fields `status` and `summary`. Fall back to per-issue `getJiraI