Pick 7 Category

Input: $ARGUMENTS

Core Principles

1. Fuzzy matching is essential. Users will say “research” when the category is “Research & Analysis.” They’ll say “decisions” when the category is “Decision & Comparison.” The matcher must handle partial, case-insensitive, and synonym-based matches.

2. Category boundaries are porous. Skills tagged in one category often serve adjacent categories. When a matched category has fewer than 7 skills, extend to adjacent categories rather than returning a short list.

3. Within-category ranking uses tier and depth. Once the category is matched, rank skills by tier score and line count (proxy for depth). The highest-tier, deepest skills rise to the top.

4. Seven gives a comprehensive view. 7 skills from one category shows the user the full range of that category’s capabilities — from core tools to specialized variants. It’s enough for category mastery, not so many as to overwhelm.

5. The category itself is informative. Part of the output should describe what this category IS and what it covers, not just list its skills. The user should understand the category’s scope.

Phase 1: Category Matching

[A] USER_INPUT: [from $ARGUMENTS]

[B] CATEGORY_MATCH:

Step 1: Load all known categories from skills.json
Step 2: Score each category against user input:
    - Exact match: 10
    - Contains user input as substring: 7
    - User input contains category as substring: 5
    - Shared significant word (excluding "and", "the", "of"): 3
    - Synonym match ("research" ↔ "analysis", "decision" ↔ "choice"): 3
Step 3: Select highest-scoring category

    MATCHED_CATEGORY: [category name]
    MATCH_CONFIDENCE: [exact / partial / fuzzy]
    MATCH_SCORE: [N]

Step 4: If no match scores above 2:
    → Ask user: "I couldn't confidently match '[input]' to a category. Available categories are: [list]. Which did you mean?"

[C] CATEGORY_DESCRIPTION:
    What this category covers: [describe in 1-2 sentences]
    Typical use cases: [list 3-4]
    Adjacent categories: [list categories that share skills/tags with this one]

Phase 2: Skill Pool and Ranking

[D] CATEGORY_POOL:

Step 1: Filter skills.json to skills in MATCHED_CATEGORY
    Pool size: [N]

Step 2: IF pool >= 7:
    Rank by:
        TIER_SCORE: tier1=10, tier2=8, category=7, tier3=4, tier4=2
        DEPTH_SCORE: line_count > 200 = 4, 100-200 = 3, 50-100 = 2, <50 = 1
        TOTAL = TIER_SCORE + DEPTH_SCORE
    Take top 7

Step 3: IF pool < 7:
    Take all from matched category
    Remaining slots = 7 - pool
    Fill from adjacent categories:
        - Skills sharing tags with matched-category skills
        - Skills invoked by matched-category skills
        - Skills in the nearest related category
    Rank adjacent candidates by relevance to matched category
    Take top [remaining slots]

[E] SELECTED: [list of 7 skills with scores]

Phase 3: Category Landscape

[F] LANDSCAPE:

Step 1: Group the 7 picks by function:
    Analysis skills: [list]
    Decision skills: [list]
    Planning skills: [list]
    Validation skills: [list]
    Other: [list]

Step 2: Identify the category's "star" skill:
    The single most-used/highest-tier/most-connected skill in this category
    STAR: /[id] — [why it's the star]

Step 3: Identify the category's "hidden gem":
    A lower-tier or less-connected skill that is surprisingly useful
    GEM: /[id] — [why it's worth knowing]

Step 4: What's NOT in this category that users often need alongside it?
    COMPLEMENT_CATEGORIES: [list 1-2 categories that pair well]

Phase 4: Output

ALGORITHM: CATEGORY
CATEGORY: [matched category]
MATCH: "[user input]" → "[matched category]" (confidence: [exact/partial/fuzzy])
POOL: [N skills in category]
PICKED: 7

CATEGORY OVERVIEW:
  [1-2 sentence description of what this category covers]
  Use cases: [list 3-4 typical scenarios]

7 SKILLS FROM [CATEGORY]:

  1. /[id] — [title] ★ [if star skill]
     [1-line description]
     Tier: [tier] | Lines: [N] | Score: [total]

  2. /[id] — [title]
     [1-line description]
     Tier: [tier] | Lines: [N] | Score: [total]

  [continue to 7...]

  [if any from adjacent categories:]
  Note: Skills [N-7] are from adjacent category "[name]" — included because [reason]

CATEGORY LANDSCAPE:
  Analysis: [skills]
  Decision: [skills]
  Planning: [skills]
  Validation: [skills]

STAR SKILL: /[id] — [why]
HIDDEN GEM: /[id] — [why]

COMPLEMENT: For skills that pair well with [category], see: [adjacent categories]
  Try: /p7cat [adjacent category name]

REMAINING: [N - 7] more skills in this category not shown
  Use /pick [N] category [name] to see more

Failure Modes

Depth Scaling

Default: 2x. These are floors.

Pre-Completion Checklist

• User input fuzzy-matched to a category with confidence level
• Category described with overview and use cases
• Exactly 7 skills returned
• Adjacent categories used if matched category is too small
• Skills ranked by tier + depth
• Star skill and hidden gem identified
• Complement categories suggested
• Remaining skills count noted

Integration

• Shortcut for: /pick 7 category $ARGUMENTS
• Use when: You know the category you want to explore
• Routes to: The 7 picked skills; /p7cat [adjacent] for related categories
• Related: /p8tier (filter by tier instead), /p10diverse (cross-category selection)
• Differs from /p8tier: cat filters by topic; tier filters by quality level
• Differs from /p10diverse: diverse spans all categories; cat goes deep in one