--- name: skill-stats description: "Use when reviewing which QE skills are being used, finding undertriggering skills, or analyzing skill effectiveness. Shows usage patterns and recommendations." user-invocable: true --- # Skill Usage Statistics View QE skill usage patterns to identify popular skills, undertriggering skills, and optimization opportunities. ## Activation ``` /skill-stats ``` ## What It Reports 1. **Most Used Skills** — Top 10 skills by invocation count 2. **Undertriggering Skills** — Skills with good descriptions but low usage 3. **Never Used Skills** — Skills that have never been triggered 4. **Trigger Method** — How skills are activated (explicit /command vs auto-selected) 5. **Recommendations** — Skills to improve, merge, or deprecate ## Usage Log Format Skill invocations are logged to `${CLAUDE_PLUGIN_DATA}/skill-usage.log`: ``` 2026-03-18T10:30:00Z|security-testing|explicit|/security-testing 2026-03-18T10:45:00Z|qe-test-generation|auto|model-selected 2026-03-18T11:00:00Z|mutation-testing|explicit|/mutation-testing ``` ## Hook Configuration ```json { "hooks": { "PreToolUse": [ { "matcher": "Skill", "hook": ".claude/skills/skill-stats/scripts/log-usage.sh" } ] } } ``` ## Analysis Script ```bash #!/bin/bash # analyze-usage.sh LOG="${CLAUDE_PLUGIN_DATA:-$HOME/.claude/plugin-data}/skill-usage.log" if [ ! -f "$LOG" ]; then echo "No usage data yet. Skills will be logged as they are used." exit 0 fi echo "=== Skill Usage Report ===" echo "" echo "Top 10 Most Used:" cut -d'|' -f2 "$LOG" | sort | uniq -c | sort -rn | head -10 echo "" echo "Trigger Method Breakdown:" cut -d'|' -f3 "$LOG" | sort | uniq -c | sort -rn echo "" echo "Skills Used in Last 7 Days:" WEEK_AGO=$(date -d '7 days ago' +%Y-%m-%dT%H:%M:%S 2>/dev/null || date -v-7d +%Y-%m-%dT%H:%M:%S) awk -F'|' -v since="$WEEK_AGO" '$1 >= since' "$LOG" | cut -d'|' -f2 | sort | uniq -c | sort -rn ``` ## Gotchas - Log file grows unbounded — rotate periodically with `tail -n 1000 $LOG > $LOG.tmp && mv $LOG.tmp $LOG` - Auto-selected skills may not always be the best match — review undertriggering skills for description improvements - Usage count alone doesn't indicate quality — a skill used 100x but failing 50% needs fixing, not celebrating