59/100 (Medium)

OpenClaw： suspicious
VirusTotal： benign
StaticScan： unknown

OpenClaw： suspicious

The skill's documentation promises real-time leaderboard fetching and integrations, but the shipped code uses simulated/local data and the examples reference external webhooks and system config change...

[内容已截断]

VirusTotal： benign VT 报告

静态扫描： unknown

README 未提供

{
    "latestVersion": {
        "_creationTime": 1772339464833,
        "_id": "k97epkensag46j4fyf0tyc1jhd822ek8",
        "changelog": "🚀 Model Benchmarks v1.0.0 - Initial Release\n\n🧠 CORE FEATURES:\n• Real-time AI capability tracking from multiple leaderboards\n• LMSYS Chatbot Arena integration (100+ models, daily updates)\n• BigCode programming leaderboard (50+ models, weekly updates)\n• HuggingFace Open LLM leaderboard (200+ models, daily updates)\n• Alpaca Eval instruction-following benchmark (80+ models)\n\n💰 COST OPTIMIZATION:\n• Performance-per-dollar calculations for all tracked models\n• 445x cost efficiency discovery (Gemini 2.0 Flash vs expensive models)\n• Task-specific model recommendations (coding, writing, analysis, translation, math, creative, simple)\n• Real-time pricing integration from OpenRouter and provider APIs\n\n📊 INTELLIGENT ANALYSIS:\n• Unified 0-100 scoring system across all capabilities\n• Multi-dimensional performance tracking (general, reasoning, creative, coding, knowledge, comprehension)\n• Trend analysis and performance change detection\n• Export capabilities for custom analysis (JSON, CSV)\n\n🔗 PERFECT INTEGRATION:\n• Seamless compatibility with model-manager skill\n• Auto-sync capabilities to compute routing systems\n• CLI and programmatic API access\n• Cross-platform Python implementation (3.8+)\n\n🎯 PROVEN RESULTS:\n• Users report 60-95% AI cost reduction\n• Data-driven model selection replaces guesswork\n• Discover hidden gem models with superior cost efficiency\n• Optimize for specific task types with intelligence\n\nFIRST RELEASE - Complete AI intelligence platform for OpenClaw optimization!",
        "changelogSource": "user",
        "createdAt": 1772339464833,
        "version": "1.0.0"
    },
    "owner": {
        "_creationTime": 0,
        "_id": "publishers:missing",
        "displayName": "Notestone",
        "handle": "notestone",
        "image": "https:\/\/avatars.githubusercontent.com\/u\/98139853?v=4",
        "kind": "user",
        "linkedUserId": "kn7858e3hv3fzgc4gydvhm25gx81pjjb"
    },
    "ownerHandle": "notestone",
    "skill": {
        "_creationTime": 1772339464833,
        "_id": "kd75qq91jk857edm96vprq9hrn8235jy",
        "badges": [],
        "createdAt": 1772339464833,
        "displayName": "AI Intelligence Hub - Real-time Model Capability Tracking",
        "latestVersionId": "k97epkensag46j4fyf0tyc1jhd822ek8",
        "ownerUserId": "kn7858e3hv3fzgc4gydvhm25gx81pjjb",
        "slug": "model-benchmarks",
        "stats": {
            "comments": 0,
            "downloads": 244,
            "installsAllTime": 1,
            "installsCurrent": 1,
            "stars": 0,
            "versions": 1
        },
        "summary": "Real-time AI model capability tracking via leaderboards (LMSYS Arena, HuggingFace, etc.) for intelligent compute routing and cost optimization",
        "tags": {
            "ai": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "benchmarks": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "bigcode": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "cost-optimization": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "data-driven": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "huggingface": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "intelligence": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "latest": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "lmsys": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "models": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "performance": "k97epkensag46j4fyf0tyc1jhd822ek8",
            "routing": "k97epkensag46j4fyf0tyc1jhd822ek8"
        },
        "updatedAt": 1772339482789
    }
}