AgentSkillsCN

requirement-prioritization

针对已明确需求的发布计划排序器。仅当`REQ-*`项目已最终敲定,且Codex需产出一份兼顾依赖关系的优先级排序与发布截断点时使用;切勿用于需求发掘。

SKILL.md
--- frontmatter
name: requirement-prioritization
description: Release planning ranker for already-defined requirements. Use only when `REQ-*` items are finalized and Codex must produce a dependency-aware priority order and release cut line; do not use for requirement discovery.

Requirement Prioritization

Trigger Boundary

  • Use when canonical REQ-* items exist and release sequencing is needed.
  • Do not use before baseline definition; use requirements-definition first.
  • Do not use for implementation decomposition; use user-story-writing.

Goal

Produce a defensible and reproducible priority order under capacity constraints.

Shared Requirements Contract (Canonical)

  • Use ../requirements-definition/references/requirements-governance-contract.md as the single schema and gate source.
  • Track requirements workflow artifacts with RQM-* IDs.
  • Run machine validation: python3 ../requirements-definition/scripts/validate_requirements_contract.py --manifest <path/to/manifest.json>.

Inputs

  • Approved REQ-* backlog
  • Effort and dependency estimates
  • RSK-* risk scores and mandatory compliance constraints
  • Capacity and release window

Outputs

  • Ranked backlog with rationale
  • Release cut line and deferred set
  • Dependency-feasible implementation order

Workflow

  1. Freeze scoring dimensions and weighting before ranking.
  2. Score each REQ-* using value, urgency, effort, and risk reduction.
  3. Apply dependency constraints to enforce executable sequence.
  4. Force mandatory compliance items above discretionary items.
  5. Publish ranked list with rationale and deferred-item triggers.

Quality Gates

  • Priority decisions are reproducible from documented rules.
  • Dependency violations are absent.
  • Compliance-blocking requirements are never deferred without escalation.
  • Ranking references source IDs and risk IDs where applicable.

Failure Handling

  • Re-score when assumptions or capacity inputs change.
  • Stop approval when mandatory legal or security requirements are deprioritized.