ggs-guess-generation-selection

name	ggs - Guess Generation → Selection
description	Generate exhaustive guesses then systematically select the top 20 for evaluation. Chains /gg into /selection.
output	{"format":"prose"}

Guess Generation → Selection

Input: $ARGUMENTS

---

Overview

This skill chains two procedures:

1. Guess Generation (/gg) — exhaustively generate guesses about the input
2. Selection (/selection) — evaluate and select from the top 20 guesses

---

Step 1: Generate Guesses

→ INVOKE: /gg $ARGUMENTS

Complete the full guess generation process. Once finished, collect all generated guesses and proceed to Step 2.

---

Step 2: Select Top 20

From the guess generation output, identify the top 20 guesses by prioritizing:

1. All CRITICAL guesses (High Impact × Low Confidence)
2. Then HIGH-CRUX guesses
3. Then highest-impact guesses from remaining pool

Compile these 20 guesses into a numbered list.

---

Step 3: Run Selection

→ INVOKE: /selection [the 20 guesses from Step 2]

Run the full selection process on these 20 guesses: triage, ARAW evaluation of critical items, dependency analysis, selection matrix, and tiered final selection.

---

Output

The final output should include:

1. Generation summary — total guesses generated, coverage metrics
2. Top 20 selected for evaluation — the numbered list fed into selection
3. Full selection output — triage buckets, ARAW evaluations, selection matrix, tiered results

---

Execution Checklist

• Full /gg completed with coverage tracking
• Top 20 guesses identified and listed
• Full /selection completed on those 20
• Tiered selections with next actions provided