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. 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.
-
kali-defense-mcp-servergithub.com/bottobot/kali-defense-mcp-serverupdated this year
Ultimate Kali Defensive Security & System Hardening MCP Server — 78 defensive security tools for firewall management, CIS benchmarks, intrusion detection, malware scanning, container security, zero-trust networking, and more.
npm MIT -
Policy-first defensive security operations server for MCP
MIT ★ 1 -
defense-mcp-claude-extensiongithub.com/defensedotcom/defense-mcp-claude-extensionupdated this year
Claude Desktop extension for querying Defense.com threat portal - view threats, security issues, and team workload
MIT ★ 3 -
aegis-secgithub.com/verrysimatupang99/aegisupdated this year
Aegis - a transparent, constitution-bound defensive security copilot with MCP server
pypi MIT -
Secure read-only MCP server for typed job search, deterministic ranking, explainable matching, adversarial defenses, and audited tool use.
MIT -
Local MCP server that helps coding agents run defensive, remediation-focused secure code review of source repositories.
npm Apache-2.0 ★ 1 -
sentricore-agent-securitygithub.com/Pothan0/sentricore-agent-securityupdated this year
Security gateway for AI agents on REST and MCP: prompt-injection defense, tool governance, human-in-the-loop escalation, tamper-evident audit logs.
-
Secure, hardened Model Context Protocol (MCP) server for safe Ubuntu system operations — policy-controlled file access, command execution, package search, and full audit trail with defense-in-depth security.
MIT ★ 1 -
Defensive security utilities for analysts and agents: IOC handling, hashing, CIDR checks, repository checks, and safe command assessment.
MIT -
bad-mcpgithub.com/canack/bad-mcpupdated this year
10 intentionally malicious MCP servers that exploit protocol features to attack AI clients. For security research and defense testing.
MIT ★ 9 -
Zero-key cybersecurity MCP server: grade any site security A+ to F, check if a CVE is actively exploited, get plain-English MITRE ATT&CK defenses. No API keys, no signups.
pypi MIT -
vpn-recongithub.com/drewburchfield/vpn-reconupdated this year
Route reconnaissance through VPN with 5-layer security defense. Claude Code plugin, MCP server, and Docker infrastructure.
MIT -
Zero-trust AI security research engine built on MCP, Tauri v2, and Rust — with prompt injection defense and sandboxed tool execution
MIT ★ 5 -
Secure MCP server for AI-assisted PCAP analysis with tshark, redaction and prompt-injection defense.
MIT -
Open-source AI security platform providing perimeter defense for LLMs and AI agents through swarm analysis, policy enforcement, adversarial testing, and real-time threat detection.
pypi AGPL-3.0 ★ 21 -
MCP-Hardenedgithub.com/Nu1lstep/MCP-Hardenedupdated this year
A deliberately small MCP server demonstrating security hardening against the OWASP MCP Top 10, with a test suite that proves the defenses hold.
MIT -
Defensive web-security auditor for applications you own: an LLM agent in a Docker sandbox, a scope guard, CVSS 3.1 findings, CI mode and an MCP server.
MIT -
@declaw/mcp-servergithub.com/declaw-ai/mcp-serverupdated this year
MCP server for Declaw — secure AI sandboxes with network policies, PII scanning, injection defense, and audit logging
npm Apache-2.0 -
Hydra-n8n-Auditorgithub.com/Owito/Hydra-n8n-Auditorupdated this year
MCP Server for agentic n8n workflow auditing and auto-patching. Implements Active Immunity and Agentic Self-Defense concepts for n8n security.
-
Security-oriented Python tooling and CLI/MCP experiments for defensive research and developer workflows.
-
Common Mitigation Enumeration (CME) — A taxonomy of defensive security controls mapped to deterministic CVSS vector attenuation, with an MCP server
★ 4 -
Security-first, read-only MCP server for Wazuh: lets AI agents investigate your SIEM while treating every log byte as hostile (prompt-injection defense, audit log, least privilege).
MIT -
A n8n community node for use with https://github.com/btafoya/spamassassin-mcp, A secure, containerized Model Context Protocol (MCP) server that integrates SpamAssassin for defensive email security analysis. This server provides Claude Code with comprehensive email analysis capabilities while maintaining strict security boundaries.
npm MIT ★ 1 -
Declaw -- Secure AI Sandboxesno update data
Secure runtime for AI agents. Run AI agents and their tools in Firecracker microVMs with network policies, PII scanning, prompt injection defense, and audit logging. 7 tools: create_sandbox, run_command, read_file, write_file, list_files, kill_sandbox, list_sandboxes. Security presets (none / standard / strict) let you control guardrails per sandbox: restrict outbound traffic to specific domains, redact PII from command output, block prompt injections, and get full audit trails.
smithery -
Free security & privacy scanner for AI-coded apps. 766 rules, 84 MCP tools, 13-layer defense, zero-trust sandbox, AI firewall. Runs locally, never sends data anywhere. Scan your AI-generated app for leaked keys, open databases, and injection holes in 5 seconds.
npm MIT ★ 7 -
AIDEFEND MCP is a local-first AI Security Defensive Assistant that brings the full AIDEFEND countermeasure library into your environment and turns static knowledge into actionable protection for LLMs and agentic AI systems — privately, securely, and on-device.
MIT ★ 18 -
Hardware-anchored DeFi MCP server (defensive security tool). GSD-planned rebuild of vaultpilot-mcp.
NOASSERTION -
simply-ops-guardgithub.com/nardovibecoding/simply-ops-guardupdated this year
Claude Code plugin - 49 hooks + 28 MCP tools for enforcement, security scanning, prompt injection defense, VPS ops, audit logging
AGPL-3.0 ★ 1 -
openresearch-mcpgithub.com/olanokhin/openresearch-mcpupdated this year
Zero-auth, security-hardened cross-domain research MCP server: web, academic, GitHub, news, weather, macro/finance, SEC, biomedical, Bluesky, YouTube. SSRF and prompt-injection defenses.
pypi Apache-2.0 ★ 12 -
Autonomous AI service monitor multi-agent pipeline (Triage, Detective, Surgeon, Validator) that monitors logs, diagnoses errors, and fixes them automatically. LangGraph orchestrator, MCP tools, Zero Trust security with 15 defense-in-depth controls.
-
as-redis-mcp-servergithub.com/zicheng-ai/redis-mcp-serverupdated this year
Read-only Redis AI Assistant for Qoder IDE - MCP Server with 8 semantic tools for key inspection, search, pattern analysis, health check, memory analysis, slowlog analysis, cache advisor, and readonly queries. Features deep-defense security: command whitelist, key prefix filter, data masking, rate limiting, and audit logging.
pypi MIT -
ramibotgithub.com/RamiBotAI/ramibotupdated this year
RamiBot v3.8.0 is a local-first AI security operations platform integrating multi-LLM support, a dynamic red/blue team skill pipeline, MCP tool orchestration, Docker terminal access, Tor proxy management, and an auto-integrated Kali-based tool server (rami-kali) for controlled, extensible offensive and defensive workflows
MIT ★ 29 -
FYP-XploitEye-Web-and-System-Vulnerabilities-using-Agents-LLMs-github.com/shoaib1522/FYP-XploitEye-Web-and-System-Vulnerabilities-using-Agents-LLMs-updated this year
XploitEye is a state-of-the-art, AI-powered multi-agent cybersecurity platform designed to automate the complete Cyber Kill Chain. By orchestrating advanced Large Language Models (LLMs) and industry-standard security tools, XploitEye provides a unified interface for both Red Team (Offensive) and Blue Team (Defensive) operations.
-
Dockerfile linter with image-size and CVE advisories
NOASSERTION -
convoyplangithub.com/cognis-digital/convoyplanupdated this year
Defense logistics route/sustainment planner computing fuel, resupply windows, and chokepoint risk from a YAML plan.
NOASSERTION -
Blitz Strike — a universal MCP security-audit toolbelt. Reconnaissance at speed. Analysis in depth. Validation before report.
npm MIT ★ 518 -
Nexargio Threat Scanner MCP Server for Autonomous AI (Zero-Hour Web Threat & Quishing Defense)
MIT -
secure-banking-mcpgithub.com/Lawaa/secure-banking-mcpupdated this year
Enterprise-grade FastMCP (Python 3.12) server demonstrating identity-isolated AI tool execution with out-of-band JWT authentication, multi-tenant database guards, and automated Pytest/Allure security testing.
-
智能体主动防御哨兵 — 文件完整性基线 / 可疑进程监控 / 监听端口告警,BIT 生态内可独立运行的 Rust 守护进程
Apache-2.0 ★ 2 -
Active Directory security diagnostics for defenders — non-privileged, read-only, Pure Rust. AS-REP/Kerberoast/LDAP/delegation findings with HTML reports, AI analysis, and MCP. For authorised assessment.
cargo MIT ★ 4 -
MISP (Malware Information Sharing Platform) MCP server with built-in prompt injection defense via prompt-defense-audit
MIT ★ 2 -
goop-shield-communitygithub.com/kobepaw/goop-shield-communityupdated this year
Runtime defense for AI agents. 24 inline defenses, 3 output scanners, MCP server, framework adapters.
Apache-2.0 ★ 2 -
CloneGuard : local release-exposure auditing, platform-aware hardening skill, MCP tools, and verifiable release receipts.
MIT -
Open-source MCP and HTTPS screening for prompts, skills, and tool descriptions using TypeSafe AI Jev, deterministic checks, and configurable reasoning.
npm ISC ★ 2 -
Aegis-SSH-MCPgithub.com/nextgridit-team/Aegis-SSH-MCPupdated this year
MCP-native SSH bridge for AI agents with host-scoped access, rule-based command validation, ephemeral SSH sessions, and audit logging
MIT ★ 4 -
Deterministic security gateway & remote MCP server for autonomous AI commerce — 12 policy rules blocking prompt-injected purchases with zero financial leak
MIT ★ 2 -
面向 Android 应用包(APK)的静态安全分析、反编译源码检索与 MCP 代码审计平台 · Android Static Security Analysis & Code Audit Platform
MIT ★ 4 -
A minimal, modular MCP server that equips your AI with practical capabilities for real-world threat hunting workflows.
NOASSERTION ★ 21 -
Anthropic MCP servers to enable AI agent integration and autonomous defense for Check Point firewalls, endpoint, and more.
MIT ★ 1 -
acrf-mcp-scangithub.com/kannasekar-alt/acrfupdated this year
MCP server inventory and risk scanner. Implements ACRF-03 defense pattern.
pypi Apache-2.0 -
unplug-mcpgithub.com/UnplugAI/unplug-mcpupdated this year
MCP server for Unplug — LLM defense layer
pypi Apache-2.0 ★ 2 -
Original open-source static software analysis toolkit: identify, hash & inspect PE/ELF/Mach-O, WebAssembly, archives (APK/JAR) and JavaScript. CLI, pluggable providers, evidence model & MCP server. Static-only, no cloud.
MIT ★ 2 -
spectyn-secopsgithub.com/markl-a/spectyn-secopsupdated this year
Multi-agent SecOps research playground — red/blue team simulation built on phantom-mesh. MCP server exposing recon/exploit/detect/respond tools to Claude Code, Cursor, or any MCP-compatible runtime.
NOASSERTION -
Protect your shipped JS from AI reverse-engineering. REA can reverse-engineer your app; REAShield returns nothing. Measured against the real REA (6 → 0 on a sample app). Not unbreakable.
npm MIT ★ 2 -
Electric Blue MCP provides a Dockerized MCP server exposing a curated set of open‑source recon and application security tools (nmap, nikto, sqlmap, dirb, dnsrecon, wpscan, searchsploit, grype, syft). All tool calls are sanitized, time‑capped, output‑truncated.SBOM and vulnerability scanning to support supply chain visibility.
★ 1 -
Antaryami is a cryptographic intent-locking tool for AI agents which acts as a middleware that mathematically proves an AI agent's sequence of actions matches the original, authenticated human intent before any external API or system command is executed. Also, ships as an MCP server for Claude Code, Cursor, or any MCP client.
MIT -
Production-grade, asynchronous Model Context Protocol (MCP) server for Claude Desktop & Cursor with Zero-Trust Read-Only SQL enforcement (AST/Regex + SQLite mode=ro), context-window bounds, and microsecond telemetry.
★ 1 -
@idevelopers/leak2lockactive
Defensive exposure-response MCP tools for founders and small developer teams.
npm MIT -
@yigitguldal/unreaactive
Defensive client-exposure scanner with an MCP server for AI coding agents.
npm MIT -
A defensive range for MCP and AI-agent attacks: reproduce a published insecure default, then ship the detection.
MIT