Mingjing— AI Agent Health Center — Mingjing— AI 代理 健康 Center

Mingjing — AI 代理 健康 Center

Paste this into ClawHub publish page. Replace screenshot URLs with actual images.

One-liner

Zero-dependency, offline-first LLM 代理 健康 监控ing. pip 安装, zero LLM calls, ~40MB RSS.

Screenshots 健康 仪表盘: [screenshot-url] Diagnosis 报告: [screenshot-url] 安装

Step 1 — 安装 backend:

pip 安装 mingjing

Step 2 — Enable probe in OpenClaw:

OpenClaw 插件s 安装 mingjing-probe OpenClaw config 设置 插件s.entries.mingjing-probe.enabled true OpenClaw gateway

Step 3 — 启动 网页 panel:

python3 -m src.ming 启动 --daemon python3 -m src.ming 网页 启动 --port 18088

Visit http://localhost:18088 to see live data.

Features Feature Description Zero dependencies Pure Python stdlib, no third-party packages Offline-first No outbound traffic, data stored locally Zero LLM cost 157 rules, pure rule engine, no 模型 calls 7 框架 adapters LangChAIn / LlamA索引 / CrewAI / OpenHands / Semantic Kernel / 代理Scope / Hermes 4-level 健康 grading 健康y / Sub-健康y / Attention / Critical, per instance 网页 仪表盘 Live event 流, triage coverage, i18n (EN/中文), dark theme Diagnostic capabilities (157 rules) Layer Count Covers 系统 20+ Memory, IO, CPU, disk, file descriptors Probe 8+ Self-检查, emit rate, buffer 健康 代理 20+ Orchestration, steps, decisions, 角色s 模型 15+ 令牌s, latency, 输出, frequency 工具 15+ Duration, errors, 输出, execution Network 6+ Connection, DNS, 状态 codes Security 5+ Injection, sensitive data leaks Memory 5+ Retrieval, storage, windowing 插件 10+ Lifecycle, loading, errors More Full docs: Mingjing on GitHub PyPI: pip 安装 mingjing 报告 issues: https://github.com/wulun811/Ming_qiankun/issues