25 tools · all MIT, all local-first

Watch your whole stack
from one agent.

MCP servers and CLIs for the SOC, the network, and the homelab. Wazuh, MISP, TheHive, LibreNMS, and Proxmox, read like a console by your AI client.

$npx -y wazuh-mcp
lidless.dev the eye does not close

Tools

Each one wires a real system into your AI client. Writes are gated.

Security / SOC9 tools · SIEM, IDS, incident response

wazuh-mcp Read your Wazuh SIEM from your agent: alerts, agents, and CVEs, read-only. v1.1.4 misp-mcp Turn MISP threat intel into answers: events, correlations, and IOC exports on demand. v1.3.6 suricata-mcp Hunt Suricata and Zeek NSM logs from your agent: alerts, flows, and C2 analytics. v2.1.2 thehive-mcp Drive TheHive incident response from your agent: cases, alerts, tasks, observables. v1.2.3 cortex-mcp Detonate an indicator and get one verdict: Cortex analyzers, confirm-gated. v1.3.3 mitre-mcp Map alerts to ATT&CK and profile threat groups without leaving your agent. v2.0.4 zeek-mcp Query and correlate Zeek and Suricata NSM logs from your agent. v3.1.2 hotwash Build IR playbooks as flowcharts and run them step by step, agent-drivable. soc-stack One command stands up a full SOC on Proxmox: Wazuh, TheHive, Cortex, MISP, Zeek, Suricata.

Threat Intelligence & OSINT4 tools · intel and OSINT

Network6 tools · NMS, automation, mesh

Homelab6 tools · hypervisor and self-hosted apps

01 / read-first

Read-only by default

Anything that changes state needs an explicit confirm. Your agent looks before it touches.

02 / local

Runs on your machine

The servers run beside your client over stdio. Credentials stay in your environment.

03 / one shape

Learn it once

Every tool is configured and gated the same way. The docs cover all of it.