- Conduid
- Marketplace
- #security
MCP servers tagged security
4.8K live MCP servers tagged "security", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with security.
H1 MCP
Read-only HackerOne Hacker API MCP server for Claude Code, Codex and opencode
OpenProcessGuard MCP
Read-only Windows process-forensics MCP server for Claude Desktop — 17 tools across 6 hunt stages, cosign-signed releases + SBOM, zero telemetry.
Leasebroker
Local-first broker issuing time-bounded, least-privilege capability leases to AI agents and MCP servers — signed, scoped, expiring; deny-by-default with audit and human veto.
Sec Shodan MCP
MCP server for Shodan - IoT search engine
Bugsafe
Share bugs, not secrets
Techtide Harness Kit
Trust contracts for AI agent skills: JSON Schema risk tiers, tool permissions, deny-by-default egress, approval gates, prompt-injection fixtures, ~28 CI validation gates. Apache-2…
Jail
A local policy shim that blocks the MCP STDIO RCE class disclosed by OX Security in April 2026
Autonomous Security Lab Assistant
Safe-by-default MCP-style assistant for scoped security lab workflows.
D8a
Blind MCP gateway: secure HTTP access to internal services without exposing data or credentials.
Mike
A Splunk-native agent that turns SOC analyst reasoning into a queryable institutional knowledge graph, accessible from SPL, a Python CLI, and a Model Context Protocol server.
Burling Action
GitHub Action: run burling AIP conformance checks and upload SARIF to code scanning
McpO2
Policy-aware MCP routing fabric for secure remote module access, capability-based orchestration, and large-context agent execution.
Simdiff
The effect layer for AI-agent firewalls — simulate an action, get a canonical effect delta. Zero-dep, deterministic, fail-closed. 5 domains (fs, sql, shell, http, solana).
Sentori
🛡️ Security scanner for AI Agents and MCP Servers — like npm audit for AI
Pypkg Audit Ptc Agent
MCP-based context engineering patterns (PTC + PTD) achieving 51% token reduction vs ReAct for LLM agent systems
Netrecon MCP
Network reconnaissance MCP server — DNS, WHOIS, port scanning, subdomain enumeration, TLS inspection, and more. Zero API keys required.
Codex Gateway MCP Server
MCP gateway wrapping Codex CLI with allowed_paths/forbidden_paths guards, command policy, and audit logging. Claude Code (planning/review) + Codex (implementation) split workflow.
Peakstar Mcpguard
MCP / Agent security scanner: scan for security holes before your AI agent touches company systems. Zero dependencies, static, runs locally, sends nothing out.
Codetrust
AI Governance Enforcement Platform — governs what AI coding agents can do, before they do it.
Agentic Deep Audit
Evidence-first repository audit engine for local & GitHub projects. Python CLI with safety model, provenance tracking, supply-chain/license/risk analysis, MCP handoff, schema vali…
DB Safe MCP
Local MCP server for safe read-only PostgreSQL access
ThreatModel
Open-source, MCP-first threat modeling that proposes and sandbox-validates the fix — not just the report. Apache-2.0."
Blindkey
Local-first credential broker for AI agents — secrets your agents can use but never see. 66 testable security constraints. Rust.
Azure Read Only
Instalador seguro do Azure MCP somente leitura para Claude Code, com skill e hooks contra operações destrutivas.
Flowki Claude Code Hardening
Deny-list config, documented attack patterns, and pentester checklist from a hands-on security review of Claude Code's permission architecture.
OpenMoat
Local security gateway for MCP with deny-by-default policy and stateless protocol support.
Proven Gate
Deterministic provenance policy layer between LLM agents and their tools (including MCP) — blocks tool calls driven by untrusted data. No LLM, no network, fail-closed.
Provekit MCP
Hardened MCP server that gives an AI agent a code security scanner, built so the tools can't be turned against the host, with a red-team suite that spawns the real server and prov…
State Of MCP Security
Reproducible security audit of MCP servers — 75% of scanned servers exposed a medium+ hardening issue to AI agents. Round 1: official reference + popular hosted servers.
Toolwall
Information-flow firewall for MCP tool calls: tracks what a session has read and blocks the call that would send it out.
Technitium Safe MCP
Constrained Technitium DNS MCP server for bounded lookups and health checks
Devdb
MCP server for safe read-only introspection of dev databases: schema, sample rows, EXPLAIN — behind a quote-aware SQL guard, column masking and query budgets. Postgres + SQLite
Scopewatch
Local-first CLI trust layer for MCP servers - shows a plain-language capability diff before every install and update
Pathguard
Go library that judges whether a path may be touched — system locations, credential stores, agent configuration and a server’s own directories are compared by file identity as wel…
Agentbouncer Js
Official JavaScript and TypeScript SDK for AgentBouncer
Shesh Audit
GuardedMCP policy gate, hash-chained audit log, kernel bridge, tool pins
Skillrank
What can this agent skill actually do to your machine? Static security grading for AI agent skills — 3,771 scanned in seconds, zero API cost.
Vetter Fixture Target
Test fixture for mcp-vetter-agent security audit demo
Skill Guard
Static security scanner for agent skill and instruction files - dotnet tool + GitHub Action. Pulled from my private toolbox and cleaned up for release.
Heldfast
You approve an MCP server. Heldfast records what you approved. Later it tells you what moved, and guard refuses the rest.
Skill Tool
AI Agent Skill 蒸馏工坊 — 从各大平台（Claude Code / Cursor / GPTs / MCP / Dify / Coze）蒸馏高质量 skill，适配 Hermes Agent 格式。技能市场和高质量 AI Agent 技能集合。
Secure Ubuntu MCP
Secure, hardened Model Context Protocol (MCP) server for safe Ubuntu system operations — policy-controlled file access, command execution, package search, and full audit trail wit…
Waqf MCP
Edge-native Islamic MCP Federation Gateway connecting Muslim developers and AI assistants to authentic Islamic sources (@IslamicSources)
2026 Quitcode 05 MCP Hw
QuitCode Workshop 5 — домашнє завдання: MCP (власний сервер LeadDesk, готові сервери з найменшими правами, A/B, threat model)
Enterprise MCP Gateway Core
Enterprise-Grade MCP Gateway with Dynamic RBAC, Multi-Backend Integration, and Defense-in-Depth Observability
Gmail Assistant Jwt
Asistente inteligente de correo electrónico para Gmail basado en el protocolo FastMCP, potenciado por un LLM backend con Ollama y protegido mediante autenticación JWT con firmado…
Surface Scan
Read an MCP client config and report what each server can actually reach — before you mount it.
Grade
Deterministic 0-100 capability-hygiene score for MCP servers: undeclared writes, unbounded params, baked-in secrets. No LLM in the loop. Zero-dep CLI + skill.
Keycloak Auth
mcp-server-keycloak-auth
Oauth Protected MCP
Oauth Protected MCP Server