From hook-hero
Query agent session metrics from hook-hero telemetry data. Use when the user asks about agent activity, efficiency, token spend, tool usage, session history, cost tracking, or productivity metrics. Examples: "how efficient was I this week", "show me my agent metrics", "which repo used the most tokens", "how much did I spend on Claude today".
How this skill is triggered — by the user, by Claude, or both
Slash command
/hook-hero:agent-metricsThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Query and analyze agent session telemetry captured by the hook-hero plugin.
Query and analyze agent session telemetry captured by the hook-hero plugin.
Session summaries: ~/.claude/hook-hero/sessions/{YYYY-MM-DD}/{session_id}.json
Raw events: ~/.claude/hook-hero/events/{YYYY-MM-DD}/{session_id}.jsonl
Use Glob to find session files, Read each one, aggregate in a Bash one-liner:
# Today's sessions
ls ~/.claude/hook-hero/sessions/$(date +%Y-%m-%d)/*.json 2>/dev/null | wc -l
grep -rl '"repo": "acme/my-app"' ~/.claude/hook-hero/sessions/
grep -rl '"channel": "claude-cli"' ~/.claude/hook-hero/sessions/
cat ~/.claude/hook-hero/events/{date}/{session_id}.jsonl
Each JSON file contains:
When presenting metrics, highlight:
npx claudepluginhub damahua/claude-code-hook-hero --plugin hook-heroCollects agent usage metrics from git history and generates health reports or dashboards. Use for measuring adoption rates, system health, or periodic reviews via Python script.
Generates daily cost reports for Claude Code usage including total spend, running sessions, token stats via claude-view MCP tools. Useful for cost, spending, budget queries.