From emdx
AI-assisted task triage — analyze all ready tasks and recommend priorities based on epic progress, dependencies, category, and age.
How this skill is triggered — by the user, by Claude, or both
Slash command
/emdx:prioritizeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Analyze all ready tasks and recommend a priority ordering.
Analyze all ready tasks and recommend a priority ordering.
Run these commands to understand the current state:
emdx task ready --json
emdx epic list
For each ready task, get full details:
emdx task view <id>
Focus on: description, epic membership, category, dependencies, age, and any work log entries.
Consider these factors when ranking tasks:
| Factor | Why it matters |
|---|---|
| Epic progress | Tasks in nearly-complete epics should be prioritized to close out the epic |
| Blocked-by chains | Tasks that unblock other tasks should be done first |
| Category | FIX > FEAT > ARCH > DOCS > CHORE (bugs before features) |
| Age | Older tasks may indicate forgotten work or growing tech debt |
| Description clarity | Well-defined tasks are easier to execute and should be prioritized |
Show the user a ranked table like:
Priority ID Title Reasoning
──────── ──────── ───────────────────────────── ──────────────────────
1 (P1) FEAT-8 Implement auto-linking Unblocks 3 downstream tasks
2 (P1) TUI-10 Wire QA presenter Nearly completes QA epic (7/8)
3 (P2) FEAT-13 Add semantic search caching Quick win, small scope
...
Priority scale:
After the user confirms (or adjusts), apply priorities:
emdx task priority <id> <1-5>
Only set priorities that differ from the default (3). Skip tasks the user wants left at normal priority.
npx claudepluginhub arockwell/emdx --plugin emdxFetches open issues from an issue tracker via MCP, enriches with historical metrics, and runs an AI priority engine to suggest fix order.
Categorizes backlog issues from Linear/GitHub or manual lists into Leverage/Neutral/Overhead, analyzes time allocation, and suggests optimizations or labels.
Plans sprints by analyzing backlog, overdue tasks, and dependencies in Astravue. Useful when organizing upcoming work or reviewing the backlog.