sentinel-agent-system
Sentinel — 自主系统健康守护者
Sentinel 是一个面向云原生环境的多智能体基础设施监控平台。除了实时调查异常之外,它还充当平台与运维遥测数据之间的自然语言接口,让管理员无需编写 SQL 即可查询基础设施健康状况。
四个专业智能体——Monitor、Diagnostician、Remediation 和 Data Intelligence——通过 LangGraph 状态机协作,并完全通过一个标准化的 MCP 网关进行协调。
架构
核心工作流是一个循环的 LangGraph 状态机:Monitor 检测异常,Diagnostician 使用 RAG 知识库确定可能的根本原因,Remediation 尝试适当的修复(遵守安全护栏),Verify 确认修复是否生效。如果验证失败,图会循环回 Diagnostician,最多达到可配置的重试上限。
所有智能体仅通过 MCP(模型上下文协议)网关与系统交互——没有智能体直接访问数据库、日志或知识库。这使得每次工具调用都可审计,并且可以在一个接口之后进行替换。
Related MCP server: dagster-mcp
特性
四个 CrewAI 智能体,每个智能体仅限访问其角色所需的 MCP 工具
基于 RAG 的诊断,针对一组精选的运维手册,使用 Sentence-Transformers 嵌入和 FAISS 相似度搜索
自然语言转 SQL 接口,面向基础设施遥测数据库,对超出范围的问题提供优雅降级
安全护栏,当系统处于关键负载时阻止高风险修复命令(例如服务重启)——已在真实条件下验证能够正确阻止,而不仅仅是理论上
自愈重试循环:失败的验证自动路由回诊断阶段,而不是静默失败
完整可观测性:通过 LangSmith 对每次 NL2SQL 调用进行追踪
技术栈
层级 | 工具 |
智能体编排 | CrewAI |
工作流 / 状态机 | LangGraph |
工具网关 | FastMCP |
LLM 推理 | Groq ( |
嵌入 + 检索 | Sentence-Transformers ( |
可观测性 | LangSmith |
数据库 | SQLite |
设置
python -m venv venv
venv\Scripts\activate # Windows
pip install -r requirements.txt创建一个 .env 文件,内容如下:
GROQ_API_KEY=your_key
LANGCHAIN_TRACING_V2=true
LANGCHAIN_API_KEY=your_langsmith_key
LANGCHAIN_PROJECT=sentinel-agent-system
运行演示
python demo.py这会针对一个示例服务器运行完整流水线,打印每个智能体在 Monitor → Diagnose → Remediate → Verify 过程中的推理,包括第一次验证失败时重试循环的实时演示。
交付物说明
Post-Mortem Report、LangSmith NL2SQL 追踪截图和工作流图已按照作业要求单独提交。
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- Alicense-qualityDmaintenanceA multi-agent MCP server that turns LLMs into an autonomous incident-response copilot, enabling rapid investigation, correlation, and remediation of production incidents.MIT
- AlicenseAqualityAmaintenanceAn MCP server that gives AI agents full visibility and control over your Dagster instance, enabling autonomous monitoring, diagnosis, and remediation of data pipelines.1710MIT
- Flicense-qualityDmaintenanceIntegrity monitor for MCP server ecosystems, providing real-time health checks, drift detection, and cascade impact analysis for any AI agent.
- AlicenseAqualityDmaintenanceAn MCP server that exposes live network monitoring data as Resources and diagnostic capabilities as Tools, letting AI assistants query network health conversationally.6MIT
Related MCP Connectors
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
Control plane for autonomous software labor. Agents claim objectives over MCP with audit trail.
Autopilot MCP server for GEO analyses, reports, content, audits, memories and agents.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/RahemShakeel/sentinel-agent-system'
If you have feedback or need assistance with the MCP directory API, please join our Discord server