jira-taskslisted

# Purpose Read Jira to answer "what am I working on", "what's the status of `<PROJECT-KEY>-115`", or to reconcile workspace tasks against tickets. **Read-only by default** — every write is a separate, explicit, user-requested action. `<jira-base-url>` and `<PROJECT-KEY>` come from the project's `CLAUDE.md` or `.claude/project.json`; if undocumented, ask once. # Rules - Every write (create / transition / comment / worklog) requires an explicit user request in the current conversation — a request to list or check tickets never authorises a mutation. Sole exception: `jira-pack`'s `/jira-fix` runs carry their own write mandate. - Resolve Atlassian MCP tools by tool name via ToolSearch, never by a hardcoded channel prefix — two channels can expose the same tools under different prefixes. - Every Jira call needs a cloudId from `getAccessibleAtlassianResources`; on any 404, re-resolve rather than retrying a stale pinned id. - Render compact tables (key, status, summary, updated), most-recently-updated first — never dump raw issue JSON. - The `Tickets:` line in a workspace task-card README is the join key both directions; adding it is a workspace edit, never a Jira write. # Resources - Read `resources/tool-flow-and-jql.md` when making any Jira call — deferred-tool loading, cloudId resolution, read-tool signatures, JQL recipes, output shape. - Read `resources/write-policy-and-task-links.md` when a write is requested or when linking tickets to workspace task cards — the full write