# Priority Engine

> Decide what should happen first when multiple tasks, ideas, or bets compete for time. Use when there are too many plausible options and the main problem is sequencing, not generation.

- Skill: `alubiama/priority-engine` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add alubiama/priority-engine`
- Raw SKILL.md: https://api.skillmd.com/api/skills/alubiama/priority-engine/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: Alubiama (https://skillmd.com/u/alubiama)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/alubiama/priority-engine

---


# Priority Engine

## Goal

Choose the best next priority under real constraints.

This skill does not generate more options. It ranks what should happen first and why.

## When To Use

- too many good options compete
- sequencing matters
- time, attention, or budget is limited
- teams argue in circles about what is "most important"

## Scope Boundaries

In scope:

- compare options
- rank by leverage, urgency, dependency, and opportunity cost
- recommend the next priority

Out of scope by default:

- portfolio optimization across everything forever
- creating options from scratch

## Escalation Conditions

Pause and keep the ranking provisional when:

- the options are not comparable on the same decision horizon
- the governing constraint is still unnamed
- key evidence gaps make ranking unstable
- sequencing depends on a bottleneck or dependency map that is still unknown

If ranking stays unstable, recommend `intent-clarifier`, `evidence-gap-detector`, or `bottleneck-locator` before deeper prioritization.

## Workflow

1. List the options under consideration.
2. Name the governing constraint.
3. Name the decision horizon.
4. Score lightly on:
   - leverage
   - urgency
   - dependency unlock
   - effort
   - opportunity cost
5. Identify any must-sequence items.
6. Define the `reprioritization_trigger`.
7. Recommend the next priority and why not the others.

## Output Contract

Always return:

1. `option_set`
2. `governing_constraint`
3. `decision_horizon`
4. `priority_logic`
5. `ranked_options`
6. `top_priority`
7. `why_not_now`
8. `reprioritization_trigger`
9. `investment_recommendation`
10. `recommended_followup`
11. `next_step`

## Guardrails

- Do not rank without naming the constraint.
- Do not confuse urgency with importance.
- If the options are not comparable yet, say so.
- Prefer a small ranked set over false precision across a long backlog.
- Revisit the ranking when the horizon or constraint changes materially.

## Anti-Patterns

- vibe-based prioritization
- ranking without trade-offs
- treating every item as top priority