MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. Each candidate is checked by a language model from its name, description and topics, and only actual MCP servers are listed (not clients, SDKs, lists or skills). Package descriptions, licences and links are read from each package's own registry (npm, PyPI, crates.io, NuGet, RubyGems; Go and Maven packages are only checked to exist), and packages the registry no longer has are removed. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
A minimal, modular MCP server that equips your AI with practical capabilities for real-world threat hunting workflows.
NOASSERTION ★ 21 -
Read-only MCP server for HEARTH, the community threat hunting hypothesis library
npm MIT -
MCP server for CrowdStrike Falcon — AI-powered threat hunting, EDR investigation, RTR, and SOC automation via Claude
pypi MIT -
Autonomous AI counterintelligence — MCP security, behavioral monitoring, memory integrity, threat hunting and forensic evidence.
-
Enterprise-grade Model Context Protocol (MCP) server for Splunk with 100 tools for search, SOC operations, threat hunting, knowledge management, and cluster diagnostics.
NOASSERTION -
Rust MCP server for Microsoft Defender XDR and Defender for Endpoint: tools for hunting, threat intelligence, vulnerability management, and gated response.
cargo MIT -
Model Context Protocol (MCP) server in Go for ThreatBook TDP (微步在线威胁检测平台): 105 OpenAPI endpoints + 4 diagnostics as typed MCP tools - alert triage, incident forensics, log hunting, threat intel & SOC automation.
MIT ★ 2 -
CrowdSentinels-AI-MCPgithub.com/thomasxm/CrowdSentinels-AI-MCPupdated this year
AI-powered threat hunting and incident response MCP server for Elasticsearch/OpenSearch
pypi GPL-3.0 ★ 204 -
All-in-one security testing toolbox that brings together popular open source tools through a single MCP interface. Connected to an AI agent, it enables tasks like pentesting, bug bounty hunting, threat hunting, and more.
MIT ★ 219 -
Connect AI agents to CrowdStrike Falcon for automated security analysis and threat hunting
go MIT ★ 282 -
🛡️ Autonomous Multi-Agent Threat Hunting & Incident Response Swarm engineered with AutoGen, Azure Sentinel, and Model Context Protocol (MCP). Sub-second Netfilter isolation & real-time KQL hunting analytics.
MIT ★ 1 -
mcp-hayabusagithub.com/omeryemba/mcp-hayabusaupdated this year
MCP server wrapping the Hayabusa Windows event log forensic analysis CLI for AI-assisted threat hunting.
MIT ★ 1 -
Enterprise autonomous phishing triage and SOC automation platform combining LLMs with the Model Context Protocol (MCP) for SIEM/EDR log correlation and threat hunting.
-
MCP server for Suricata IDS/IPS and Zeek NSM: analyze EVE JSON alerts, hunt threats, and manage rules from an AI client
npm MIT ★ 4 -
Production-grade MCP server for Wazuh SIEM — 55 security tools for alert triage, threat hunting, vulnerability management, compliance (PCI DSS, GDPR, HIPAA, NIST CSF, ISO 27001) and active response. Connect Claude or any LLM to your SOC. OAuth 2.1, RBAC, multi-cluster, air-gap ready.
go MIT ★ 272 -
AI-Powered Threat Intelligence MCP tool
pypi ★ 40 -
Self-hosted threat intelligence and IOC management platform
Apache-2.0 -
Natural language interface to OpenCTI threat intelligence. Built with Claude Code for $22. Part of Cooper Cyber Coffee.
NOASSERTION ★ 29 -
Execute some KQL with your favorite LLM agent.
MIT -
@iflow-mcp/cyreslab-ai-urlhaus-mcp-servergithub.com/iflow-mcp/cyreslab-ai-urlhaus-mcp-serverupdated this year
MCP server for accessing URLhaus malicious URL database from abuse.ch - provides threat intelligence for cybersecurity research
npm MIT -
splunk-soc-labgithub.com/mervintab/splunk-soc-labupdated this year
AI-powered SOC lab with Splunk Enterprise and Claude MCP integration
MIT -
MCP servers, and the typed tool clients they wrap, for security operations
Apache-2.0 ★ 1 -
Wazuh_Servidor_MCPgithub.com/nks1097/Wazuh_Servidor_MCPupdated this year
Servidor MCP (Model Context Protocol) para Wazuh SIEM com 78 ferramentas de segurança em Português e orquestração SOAR.
MIT -
MCP server for the F6 XDR platform — gives AI assistants access to alerts, incidents, mail, files and hosts
pypi Apache-2.0 ★ 3 -
Self-hosted SIEM for homelabs. SIEMBOX finds new log sources, onboards them automatically, parses raw logs, and fires alerts from detection rules. A built-in MCP server lets AI assistants search logs, review alerts, and check parse coverage.
MIT -
MCP server for AlienVault OTX / LevelBlue Open Threat Exchange — complete DirectConnect API coverage (29 tools), stdio + streamable HTTP transports, read-only by default. Built on MCP SDK 2.x
pypi MIT ★ 2 -
puck-scoutgithub.com/puck-security/puck-scoutupdated this year
Autonomous, read-only endpoint investigation via MCP. Ask a question about your fleet, get a narrative answer with containment recommendations.
MIT ★ 23 -
Read-only MCP server for Azure Log Analytics, Microsoft Sentinel and Azure Resource Graph queries.
npm MIT -
cybersec-mcpgithub.com/xu-c0/cybersec-mcpupdated this year
MCP server with 323 cybersecurity prompts and 7 chained workflows for Claude, Cursor, and other MCP-compatible clients
npm NOASSERTION ★ 1 -
Local vs. cloud LLMs triaging real Splunk alerts through the Splunk MCP Server, evaluated against manual ground truth, with a least-privilege AI identity.
-
Protocol-SIFT-Async-Bridgegithub.com/mwarsss/Protocol-SIFT-Async-Bridgeupdated this year
Type-safe async MCP server for LLM-assisted memory forensics via Volatility 3: closes the 60-second attacker breakout window without LLM timeouts or evidence spoliation.
MIT ★ 1 -
neo-finds-evilgithub.com/NikVir/neo-finds-evilupdated this year
Read-only Neo4j graph-correlation layer extending Protocol SIFT for AI-driven DFIR — cross-host, cross-time evidence correlation exposed to a Claude Code agent via architecturally read-only MCP tools. SANS FIND EVIL! 2026 submission.
Apache-2.0 -
Run a real SOC investigation on AWS — one command, two hours, an AI senior analyst at the keyboard with you. Wazuh SIEM + MCP server + Claude Code, deployable from scratch even if you've never touched any of it.
MIT -
AI-driven DFIR framework: an MCP server that turns Claude Code into an autonomous digital-forensics analyst on SANS SIFT - correlates disk + memory evidence, enforces a deterministic evidence-provenance gate, and produces auditable, hash-chained investigation reports.
MIT ★ 5 -
Yomi is engineered to operate on a fundamentally faster timeline. By orchestrating SANS SIFT Workstation forensic tools through a strict, type-safe Model Context Protocol (MCP) server and evaluating evidence via a cascading Epistemic Doubt Engine, Yomi achieves a Time-to-Containment (TTC) of < 5 seconds.
MIT ★ 2