WN - What’s New

Input: $ARGUMENTS

Core Principles

1. New means new. Not modified, not improved, not unchanged — genuinely NEW additions that didn’t exist before. The skill must distinguish new items from changes to existing items.

2. View determines grouping. The same set of new items tells different stories depending on how you group them: by time (what happened when), by category (what kind of things were added), by component (where in the system), or by theme (what pattern connects them).

3. Recency bias is real. Recent additions feel more important than they are. The skill should present new items neutrally, not weighted by how recently they appeared.

4. Context determines scope. “What’s new” in a 3-day sprint is different from “what’s new” in a 6-month project. The time range must be explicit.

5. New items have different sizes. A new architecture pattern and a new configuration option are both “new” but vastly different in significance. The output should make relative significance visible.

Phase 1: Scope Definition

[W1] PROJECT: [what project or system]
[W2] TIME_RANGE: [from when to when — or "since last review"]
[W3] SOURCE: [where to look — git log, file system, conversation history, notes]
[W4] VIEW: [by_time | by_category | by_component | by_theme | by_significance]

Phase 2: New Item Discovery

Scan the source for genuinely new additions:

[W-N] NEW: [item description]
  ADDED: [when]
  CATEGORY: [what kind of thing]
  COMPONENT: [where in the system]
  SIGNIFICANCE: [major | moderate | minor]
  EVIDENCE: [how we know this is new, not modified]

New vs Modified

Phase 3: Grouping

Group new items by the selected view:

By Time

[W-N] DAY/PERIOD: [date range]
  - [item] — SIGNIFICANCE: [level]
  - [item] — SIGNIFICANCE: [level]

By Category

[W-N] CATEGORY: [category name]
  - [item] — ADDED: [when] — SIGNIFICANCE: [level]
  - [item] — ADDED: [when] — SIGNIFICANCE: [level]

By Component

[W-N] COMPONENT: [component name]
  - [item] — CATEGORY: [type] — SIGNIFICANCE: [level]

By Significance

[W-N] MAJOR:
  - [item] — [what it is] — [when added]
MODERATE:
  - [item] — [what it is] — [when added]
MINOR:
  - [item] — [what it is] — [when added]

Phase 4: Summary

[W-N] SUMMARY:
  TOTAL_NEW: [N items]
  MAJOR: [N items]
  MODERATE: [N items]
  MINOR: [N items]
  MOST_ACTIVE_CATEGORY: [category with most new items]
  MOST_ACTIVE_PERIOD: [time period with most additions]
  PATTERN: [any visible pattern in what's being added — optional]

Phase 5: Output

WHAT'S NEW
==========

PROJECT: [name]
PERIOD: [time range]
VIEW: [grouping method]

[grouped items per selected view]

SUMMARY:
  [N] new items ([N] major, [N] moderate, [N] minor)
  Most active: [category/period]
  Pattern: [if any]

NOT INCLUDED:
  - [N] modified items (use /alebc for changes)
  - [N] unchanged items

READY FOR:
- /alebc — to see big changes (not just new items)
- /iterate [component] — to improve a specific area
- /evaluate [new item] — to assess quality of something new

Failure Modes

Depth Scaling

Default: 2x. These are floors.

Pre-Completion Checklist

• Time range explicitly defined
• Source specified and scanned
• New items distinguished from modified items
• Significance assessed for each item
• Items grouped by selected view
• Summary with counts and patterns
• Modified items excluded (noted for reference)

Integration

• Complementary: /alebc (big changes — includes modifications, not just new items)
• Differs from /alebc: alebc shows big changes; wn shows all new additions regardless of size
• Use from: project reviews, sprint retrospectives, onboarding (“what’s new since I last looked”)
• Routes to: /evaluate, /iterate for assessing or improving new items