MCP HubMCP Hub
Zurück zu Fähigkeiten

usage-tracker-3-usage-summary-reports

vamseeachanta
Aktualisiert Yesterday
9 Ansichten
3
2
3
Auf GitHub ansehen
Anderegeneral

Über

Diese Fähigkeit erstellt menschenlesbare Nutzungszusammenfassungen aus Claude-API-Protokollen, organisiert Daten nach Modelltyp (Opus, Sonnet, Haiku) und Zeiträumen. Sie liefert formatierten Output mit farbcodierten Statistiken für eine schnelle Analyse der API-Nutzungsmuster. Nutzen Sie dies, wenn Sie die Claude-Modellnutzung über verschiedene Zeiträume hinweg überwachen und berichten müssen.

Schnellinstallation

Claude Code

Empfohlen
Primär
npx skills add vamseeachanta/workspace-hub
Plugin-BefehlAlternativ
/plugin add https://github.com/vamseeachanta/workspace-hub
Git CloneAlternativ
git clone https://github.com/vamseeachanta/workspace-hub.git ~/.claude/skills/usage-tracker-3-usage-summary-reports

Kopieren Sie diesen Befehl und fügen Sie ihn in Claude Code ein, um diese Fähigkeit zu installieren

Dokumentation

3. Usage Summary Reports (+1)

3. Usage Summary Reports

Generate human-readable reports:

#!/bin/bash
# ABOUTME: Usage summary report generation
# ABOUTME: Pattern from check_claude_usage.sh

# Colors
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
BLUE='\033[0;34m'
CYAN='\033[0;36m'
NC='\033[0m'

# Generate usage summary
generate_summary() {
    local period="${1:-today}"

    # Get counts
    local opus_count=$(count_by_item "$period" "model" | grep opus | awk '{print $2}')
    local sonnet_count=$(count_by_item "$period" "model" | grep sonnet | awk '{print $2}')
    local haiku_count=$(count_by_item "$period" "model" | grep haiku | awk '{print $2}')

    opus_count=${opus_count:-0}
    sonnet_count=${sonnet_count:-0}
    haiku_count=${haiku_count:-0}

    local total=$((opus_count + sonnet_count + haiku_count))

    if [[ $total -eq 0 ]]; then
        echo -e "${YELLOW}No usage recorded for $period${NC}"
        return
    fi

    # Calculate percentages
    local opus_pct=$((opus_count * 100 / total))
    local sonnet_pct=$((sonnet_count * 100 / total))
    local haiku_pct=$((haiku_count * 100 / total))

    # Display report
    echo -e "${CYAN}═══════════════════════════════════════${NC}"
    echo -e "${CYAN}  Usage Summary - ${period}${NC}"
    echo -e "${CYAN}═══════════════════════════════════════${NC}"
    echo ""
    echo -e "  Total tasks: ${BLUE}${total}${NC}"
    echo ""
    echo -e "  ${GREEN}Opus:${NC}   ${opus_count} tasks (${opus_pct}%)"
    echo -e "  ${BLUE}Sonnet:${NC} ${sonnet_count} tasks (${sonnet_pct}%)"
    echo -e "  ${YELLOW}Haiku:${NC}  ${haiku_count} tasks (${haiku_pct}%)"
    echo ""

    # Recommendations
    if [[ $sonnet_pct -gt 50 ]]; then
        echo -e "${RED}⚠️  High Sonnet usage (${sonnet_pct}%)${NC}"
        echo -e "   Consider shifting tasks to Opus or Haiku"
        echo ""
    fi

    if [[ $haiku_pct -lt 20 ]]; then
        echo -e "${YELLOW}ℹ️  Low Haiku usage (${haiku_pct}%)${NC}"
        echo -e "   Opportunity to use Haiku for quick tasks"
        echo ""
    fi

    echo -e "${CYAN}═══════════════════════════════════════${NC}"
    echo -e "  ${GREEN}Target Distribution:${NC}"
    echo -e "  Opus: 30% | Sonnet: 40% | Haiku: 30%"
    echo -e "${CYAN}═══════════════════════════════════════${NC}"
}

4. Threshold Monitoring

Monitor against limits and thresholds:

#!/bin/bash
# ABOUTME: Threshold monitoring with alerts
# ABOUTME: Check usage against limits

# Thresholds
declare -A THRESHOLDS=(
    ["opus_daily"]=50
    ["sonnet_daily"]=100
    ["haiku_daily"]=200
    ["total_daily"]=250
    ["warning_pct"]=70
    ["critical_pct"]=90
)

# Check single threshold
check_threshold() {
    local current="$1"
    local limit="$2"
    local name="$3"

    if [[ $limit -eq 0 ]]; then
        return 0
    fi

    local pct=$((current * 100 / limit))

    if [[ $pct -ge ${THRESHOLDS[critical_pct]} ]]; then
        echo -e "${RED}⚠️  CRITICAL: $name at ${pct}% (${current}/${limit})${NC}"
        return 2
    elif [[ $pct -ge ${THRESHOLDS[warning_pct]} ]]; then
        echo -e "${YELLOW}⚠️  WARNING: $name at ${pct}% (${current}/${limit})${NC}"
        return 1
    else
        echo -e "${GREEN}✓ $name: ${pct}% (${current}/${limit})${NC}"
        return 0
    fi
}

# Check all thresholds
check_all_thresholds() {
    local period="${1:-today}"
    local status=0

    echo -e "${CYAN}Checking usage thresholds...${NC}"
    echo ""

    local opus=$(count_by_item "$period" "model" | grep opus | awk '{print $2}')
    local sonnet=$(count_by_item "$period" "model" | grep sonnet | awk '{print $2}')
    local haiku=$(count_by_item "$period" "model" | grep haiku | awk '{print $2}')
    local total=$((${opus:-0} + ${sonnet:-0} + ${haiku:-0}))

    check_threshold "${opus:-0}" "${THRESHOLDS[opus_daily]}" "Opus" || status=$?
    check_threshold "${sonnet:-0}" "${THRESHOLDS[sonnet_daily]}" "Sonnet" || status=$?
    check_threshold "${haiku:-0}" "${THRESHOLDS[haiku_daily]}" "Haiku" || status=$?
    check_threshold "$total" "${THRESHOLDS[total_daily]}" "Total" || status=$?

    return $status
}

GitHub Repository

vamseeachanta/workspace-hub
Pfad: .claude/skills/_core/bash/usage-tracker/3-usage-summary-reports

Verwandte Skills

algorithmic-art

Meta

This Claude Skill creates original algorithmic art using p5.js with seeded randomness and interactive parameters. It generates .md files for algorithmic philosophies, plus .html and .js files for interactive generative art implementations. Use it when developers need to create flow fields, particle systems, or other computational art while avoiding copyright issues.

Skill ansehen

subagent-driven-development

Entwicklung

This skill executes implementation plans by dispatching a fresh subagent for each independent task, with code review between tasks. It enables fast iteration while maintaining quality gates through this review process. Use it when working on mostly independent tasks within the same session to ensure continuous progress with built-in quality checks.

Skill ansehen

executing-plans

Design

Use the executing-plans skill when you have a complete implementation plan to execute in controlled batches with review checkpoints. It loads and critically reviews the plan, then executes tasks in small batches (default 3 tasks) while reporting progress between each batch for architect review. This ensures systematic implementation with built-in quality control checkpoints.

Skill ansehen

cost-optimization

Andere

This Claude Skill helps developers optimize cloud costs through resource rightsizing, tagging strategies, and spending analysis. It provides a framework for reducing cloud expenses and implementing cost governance across AWS, Azure, and GCP. Use it when you need to analyze infrastructure costs, right-size resources, or meet budget constraints.

Skill ansehen