How this skill is triggered — by the user, by Claude, or both
Slash command
/skill-tracker:skill-statsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run the skill tracker analysis script to display usage statistics.
Run the skill tracker analysis script to display usage statistics.
python3 "${CLAUDE_PLUGIN_ROOT}/skills/skill-stats/scripts/analyze.py" --format table
Pass through any arguments the user provides via $ARGUMENTS:
--project <path> — filter by project directory--session <id> — filter by session ID--since <YYYY-MM-DD> — filter entries since date--format json — machine-readable JSON outputExample: if the user says "/skill-stats --project /path/to/project", run:
python3 "${CLAUDE_PLUGIN_ROOT}/skills/skill-stats/scripts/analyze.py" --project /path/to/project --format table
Present the output clearly. If no data exists yet, explain that skills need to be used first for tracking data to accumulate.
npx claudepluginhub l1nusb/skill-tracker --plugin skill-trackerGenerates usage reports for Claude Code skills from tracked JSONL data. Provides CLI reports and HTML visualizations (bar charts + heatmaps) filtered by time period.
Audits installed skills by combining token cost and usage tracking into a single report, sorted by waste. Use when asking 'are my skills worth it' or to identify dead weight.
Analyzes and optimizes Agent Skills (SKILL.md) using session data and static analysis. Diagnoses underperforming skills, wasted context, and CSO violations.