MCP server category
Security MCP servers: comparison and full list
The Security category lists 133 open-source MCP servers, checked on GitHub. The best scored on our grid are Semgrep, World Monitor, and SafeDep Vet MCP. 78% received a commit in the last three months.
133servers
78 %active (3 months)
24official
19stars (median)
What Security MCP servers are for
Security MCP servers connect an AI agent to code scanning, reverse engineering, or information gathering tools. They can support developers, analysts, and people examining systems within an authorized scope. Semgrep scans source code for security vulnerabilities. GhidraMCP exposes capabilities such as decompilation and binary exploration through Ghidra. You might ask: “Analyze this authorized repository, show where each alert occurs, and separate observed findings from assumptions.” Other integrations focus on different evidence, so the server should match the material being examined. Its tools and permissions determine whether the agent can only inspect results or also modify analysis artifacts.
Example uses:
- Scan an authorized repository and review security alerts produced by the rules available to the tool.
- Inspect a binary’s imports and functions within an environment specifically reserved for analysis work.
- Decompile an authorized program to prepare an annotated review of its apparent behavior and structure.
- Consult open sources to document an investigation while treating retrieved material as untrusted evidence.
The category in figures
Of the 133 servers in the category (4% of the directory), 24 are official vendor implementations, 89% declare a licence, 17 offer a remote server and 19 need at least one API key or token according to the MCP registry. The median repository has 19 stars.
All Security MCP servers
Click a header to sort. Score out of 100: adoption, maintenance and framework (licence, registry, official status); risk: system access, secrets, licence, abandonment.
| Server | Stars | Last commit | Licence | Risk | Score |
|---|---|---|---|---|---|
| World Monitor Real-time global intelligence across conflict, markets, energy, trade, shipping and cyber. 95 tools. | 89,414 | Oct 24, 2026 | AGPL-3.0 | low | 91 |
| Semgrep ◆ Scan code for security vulnerabilities using Semgrep. | 17,241 | Oct 23, 2026 | LGPL-2.1 | low | 95 |
| Ida pro MCP IDA Pro plugin for binary analysis: decompilation, disassembly and automatic malware analysis reports. | 13,126 | Sep 26, 2026 | MIT | low | 85 |
| GhidraMCP (LaurieWired) Autonomous reverse engineering with Ghidra: decompile binaries, rename methods and data, and list methods, classes, imports and exports. | 11,510 | Oct 10, 2026 | Apache-2.0 | low | 85 |
| Robin: AI Powered Dark Web OSINT Dark web OSINT over Tor: search onion engines, scrape pages, report with your own model. | 7,660 | Oct 3, 2026 | MIT | low | 84 |
| Jadx ai MCP JADX decompiler plugin providing live, LLM-assisted reverse engineering. | 2,933 | Oct 19, 2026 | Apache-2.0 | low | 80 |
| Power BI Authoring MCP Server (Local) The Power BI Authoring MCP server lets an AI agent create and change Power BI semantic models. | 1,247 | Sep 29, 2026 | MIT | low | 82 |
| SafeDep Vet MCP ◆ Locally check npm and PyPI packages, such as those suggested by AI coding tools, for vulnerabilities and malicious code. | 1,117 | Oct 24, 2026 | Apache-2.0 | low | 90 |
| Skylos Dead code detection, security scanning and code quality analysis for Python, TypeScript and Go, with AI-assisted remediation. | 849 | Oct 24, 2026 | Apache-2.0 | moderate | 79 |
| Apktool MCP server Automate reverse engineering of Android APKs with Apktool. | 678 | Jul 2, 2026 | Apache-2.0 | low | 61 |
| SonarQube MCP Server Analyze code quality and security with SonarQube Server or Cloud directly in AI assistants. | 671 | Oct 23, 2026 | — | high | 63 |
| Emilia Protocol Requires a named human's approval before irreversible agent actions such as payments or deploys, then issues an offline-verifiable Ed25519 t | 598 | Oct 19, 2026 | Apache-2.0 | moderate | 78 |
| Binary ninja MCP Binary Ninja plugin and bridge for automating binary analysis and reverse engineering. | 456 | Oct 21, 2026 | GPL-3.0 | low | 72 |
| Wireshark MCP Wireshark packet analysis: capture, protocol statistics, field extraction and security analysis. | 353 | Oct 24, 2026 | MIT | low | 75 |
| Ori Mnemos Persistent memory infrastructure for AI agents. Identity, knowledge graph, and search. | 333 | Oct 1, 2026 | Apache-2.0 | low | 70 |
| R2mcp Disassemble and inspect binaries for reverse engineering with the Radare2 disassembler. | 331 | Oct 15, 2026 | MIT | low | 70 |
| Emisar Let AI operate servers without SSH. Choose actions, approve risky changes, and audit every step. | 330 | Oct 23, 2026 | — | moderate | 66 |
| DashClaw ◆ Fail-closed approval layer for unattended agent runs: checks actions against org policy, requests human approval and logs every decision to | 313 | Oct 21, 2026 | MIT | moderate | 85 |
| CrowdStrike Falcon MCP Server Connects AI agents with CrowdStrike Falcon for security analysis and automation. | 273 | Oct 22, 2026 | MIT | moderate | 69 |
| OpenWorkProof Verifiable execution protocol for agent tool calls: signed policy decisions, causal evidence chains and offline verification from a SQLite l | 262 | Sep 12, 2026 | Apache-2.0 | low | 69 |
| Vibe Prospecting B2B company and contact data for lead lists, enrichment, research, and outreach. OAuth sign-in. | 249 | Sep 15, 2026 | MIT | low | 64 |
| MCP server wazuh Access real-time security alerts and event data from Wazuh SIEM. | 240 | Dec 12, 2025 | MIT | low | 49 |
| PyMemoryEditor Read, write and scan the memory of running processes, Cheat Engine style: value scans that narrow down as the value changes, pointer chains, | 226 | Oct 12, 2026 | MIT | low | 74 |
| Secops MCP Security testing toolbox that combines open source tools behind a single interface for pentesting, bug bounty hunting and threat hunting. | 220 | Sep 17, 2025 | MIT | moderate | 40 |
| Reversecore MCP Security-first MCP server for reverse engineering, malware analysis, forensics, and SAST. | 212 | Oct 22, 2026 | MIT | low | 68 |
| Rsigma ◆ Author, lint, validate and convert Sigma detection rules with RSigma, evaluate and explain detections against log events, and inspect correl | 174 | Oct 22, 2026 | MIT | low | 77 |
| GhidraMCP (13bm) Ghidra plugin for binary analysis: function inspection, decompilation, memory exploration and import/export analysis. | 149 | Oct 22, 2026 | Apache-2.0 | low | 67 |
| Code Pathfinder Code intelligence MCP server: call graphs, type inference, and symbol search for Python/Go. | 143 | Jul 10, 2026 | Apache-2.0 | low | 61 |
| Shellward Agent security middleware with layered defenses: prompt injection detection, DLP data flow tracking, command blocking and PII detection. | 139 | Sep 28, 2026 | Apache-2.0 | low | 71 |
| Cve search MCP Query the CVE-Search API: browse vendors and products, get CVEs by ID and list the latest updated CVEs. | 108 | Jul 26, 2025 | MIT | moderate | 37 |
| OPNSenseMCP Manage and interact with the OPNsense open source firewall in natural language. | 92 | Sep 21, 2026 | MIT | low | 60 |
| Agentgate Read-only evidence for the tools agents run: MCP server records, coverage and policy. | 91 | Oct 24, 2026 | AGPL-3.0 | low | 65 |
| Odoo MCP Pro Connect AI to Odoo ERP 14-20: search, create, update and manage records in natural language. | 88 | Oct 23, 2026 | — | moderate | 54 |
| DocSentinel Automate cybersecurity assessment of documents, questionnaires and reports to surface risks, compliance gaps and remediations, using a RAG k | 88 | Oct 22, 2026 | MIT | low | 64 |
| Cybersec toolkit Install a large security toolset and discover and run it through an authorization-gated server for CTF, pentesting, bug bounty and DFIR. | 80 | Oct 23, 2026 | MIT | low | 69 |
| Dabble Me Journal Private journal MCP server to search, analyze, and create Dabble Me entries securely via OAuth. | 80 | Oct 24, 2026 | MIT | low | 64 |
| Dependency management MCP server Sonatype component intelligence: versions, security analysis, and Trust Score recommendations | 74 | Jan 14, 2026 | — | moderate | 40 |
| Cloud audit AWS security scanner with attack chain detection, breach cost estimation and remediation snippets for CLI and Terraform. | 73 | Oct 14, 2026 | MIT | low | 69 |
| Mobb vibe shield MCP ◆ Mobb Vibe Shield identifies and remediates vulnerabilities in human and AI-written code. | 68 | Aug 31, 2026 | MIT | low | 68 |
| MCP ZAP Server Self-hosted OWASP ZAP integration with guided security scans, findings summaries and report generation. | 67 | Oct 23, 2026 | Apache-2.0 | moderate | 68 |
| MCP Dandan Real-time security framework for MCP servers that detects and blocks malicious agent behavior by analyzing tool call patterns and intent. | 67 | Oct 22, 2026 | MIT | low | 63 |
| Kastell Server security auditing and hardening (SSH, firewall, Docker, TLS) with CIS/PCI-DSS/HIPAA mapping, fleet management and forensic evidence c | 60 | Oct 6, 2026 | Apache-2.0 | moderate | 68 |
| Studio MCP ◆ Embeds Snyk's security engines into agentic workflows to secure AI-generated code in real time and work through vulnerability backlogs faste | 58 | Oct 21, 2026 | Apache-2.0 | low | 78 |
| MCP security audit Audit npm package dependencies for security vulnerabilities with real-time checks against the npm registry. | 57 | Jul 18, 2025 | MIT | moderate | 35 |
| Roadrecon MCP server Analyze ROADrecon gather results from Azure tenant enumeration. | 53 | Mar 30, 2025 | MIT | moderate | 34 |
| MCP Security Query the ORKL API to fetch threat reports, analyze threat actors and retrieve intelligence sources. | 52 | Jan 22, 2025 | — | high | 24 |
| Volatility MCP Memory forensics with Volatility 3, exposing plugins such as pslist and netscan through REST APIs. | 52 | Jul 9, 2025 | Apache-2.0 | moderate | 34 |
| MCP dnstwist MCP server for dnstwist: detect typosquatting, phishing and brand impersonation domains. | 50 | Oct 13, 2026 | MIT | low | 62 |
| Database sentinel Read-only security audit for Supabase. Four tools surface RLS gaps, permissive policies, exposed functions and grants; an audit prompt retur | 50 | Oct 6, 2026 | MIT | low | 62 |
| CodeInspectus Local security scanner for AI-generated JS/TS that combines Opengrep, Gitleaks and Trivy with checks for exposed secrets, Supabase RLS and p | 49 | Oct 3, 2026 | Apache-2.0 | low | 67 |
| Authenticator MCP Securely interact with the Authenticator App. | 47 | Mar 31, 2026 | MIT | low | 42 |
| MCP panther ◆ Use Panther's SIEM platform in natural language to write detections, query logs and manage alerts. | 47 | Oct 23, 2026 | Apache-2.0 | low | 72 |
| Cyberchef api MCP server Use CyberChef operations through the CyberChef server API. | 44 | Apr 3, 2026 | MIT | low | 41 |
| Entraid MCP server Microsoft Entra ID (Azure AD) directory, user, group, device, sign-in and security operations via Microsoft Graph. | 41 | May 2, 2025 | — | high | 23 |
| Osv MCP Query the OSV (Open Source Vulnerabilities) database by package version or commit, batch-query multiple packages and get vulnerability detai | 41 | Oct 23, 2026 | Apache-2.0 | low | 61 |
| ContrastAPI Security intelligence: CVE/EPSS/KEV lookups, domain recon (DNS, WHOIS, SSL, subdomains), IOC threat intel, OSINT and code scanning for secre | 35 | Oct 21, 2026 | MIT | low | 71 |
| MCP recon Conversational recon with httpx and asnmap: domain analysis, security header inspection, certificate analysis and ASN lookups at multiple re | 31 | Apr 22, 2025 | — | high | 22 |
| CTRLRun Operator Execution safety for agent actions: a YAML-policy gateway that allows, holds or denies tool calls, plus an approval queue and hash-chained r | 27 | Oct 22, 2026 | Apache-2.0 | low | 64 |
| CyberChef MCP GCHQ CyberChef's encryption, encoding, compression and forensics operations, plus analysis tools such as hash identification, cipher solving | 27 | Oct 21, 2026 | GPL-3.0 | low | 64 |
| Csl core Deterministic AI safety policy engine with Z3 formal verification: write, verify and enforce machine-verifiable constraints for AI agents. | 27 | Oct 5, 2026 | Apache-2.0 | low | 59 |
| Operant MCP Security testing tools for penetration testing, network forensics, memory analysis and vulnerability assessment. | 26 | Apr 1, 2026 | MIT | low | 44 |
| Intruder MCP Access Intruder to identify, understand and fix security vulnerabilities in your infrastructure. | 26 | Oct 13, 2026 | BSD-3-Clause | low | 59 |
| Warden MCP Manage Bitwarden and Vaultwarden vaults via the bw CLI: search, create, edit and organize logins, notes, cards, identities, SSH keys, folder | 24 | Oct 24, 2026 | MIT | low | 59 |
| AIM Guard MCP Security-focused safety guidelines and content analysis for AI agents. | 23 | Oct 13, 2025 | ISC | moderate | 31 |
| Mobsf MCP server Static and dynamic analysis of Android and iOS apps with MobSF. | 23 | Sep 9, 2026 | MIT | low | 54 |
| Mastyf.ai Runtime security proxy for MCP that blocks prompt injection, SSRF, shell/SQL injection and credential exfiltration, plus trust scores for np | 21 | Oct 21, 2026 | AGPL-3.0 | moderate | 63 |
| Agentward MCP proxy that enforces least-privilege YAML policies on tool calls, classifies PII/PHI, detects dangerous skill chains and generates compli | 19 | Jul 8, 2026 | — | moderate | 36 |
| Aegis — AI Agent Governance Policy-based governance for agent tool calls across LangChain, OpenAI, Anthropic and MCP: YAML policies, approval gates, risk assessment and | 18 | Oct 16, 2026 | MIT | low | 63 |
| MCP Hangar Self-hosted policy enforcement for MCP server fleets: deterministic admission and egress policies, attributable audit logs and SIEM export. | 18 | Oct 24, 2026 | MIT | low | 69 |
| Fida Local-first MCP gateway for coding agents that redacts detected secrets from file reads and command output before they reach model context. | 18 | Jun 28, 2026 | MIT | low | 46 |
| MCP server cortex Cortex integration for observable analysis and automated security responses. | 17 | Dec 6, 2025 | MIT | low | 37 |
| MCP vms Retrieve live and recorded video from a CCTV recording program (VMS) and control it, such as opening live or playback dialogs for specific c | 17 | Aug 12, 2026 | MIT | low | 52 |
| MCP server thehive TheHive integration for collaborative security incident response and case management. | 16 | Dec 6, 2025 | MIT | low | 37 |
| DROS VajraClaw Hacker Deterministic execution governance gateway and W3C DID security guardrail for AI agent MCP tool calls. | 16 | Oct 14, 2026 | — | moderate | 47 |
| Platform Governance proxy that wraps MCP servers with policy evaluation, human approval workflows and hash-chain audit trails. | 15 | Oct 19, 2026 | Apache-2.0 | moderate | 62 |
| Quantakrypto pqc tools Post-quantum readiness: scan code for quantum-vulnerable cryptography (RSA, ECDH, ECDSA, DH), get ML-KEM/ML-DSA/SLH-DSA migration guidance a | 15 | Oct 14, 2026 | Apache-2.0 | low | 62 |
| S gw Local credential gateway for coding agents: agents get handles instead of raw secrets, and one-time approvals inject credentials only into a | 15 | Oct 24, 2026 | Apache-2.0 | low | 62 |
| AI Runtime Guard Runtime policy enforcement for AI agents that prevents accidental system damage and unauthorized access, with automatic backups before file | 15 | May 31, 2026 | MIT | low | 45 |
| Sint protocol MCP governance proxy with capability tokens, tiered approvals, fail-closed execution and tamper-evident audit receipts, plus preflight tool- | 15 | Oct 24, 2026 | Apache-2.0 | low | 57 |
| Depguard Pre-install npm package checks: static analysis, supply-chain attack detection, vulnerability audits, AI hallucination guard and CycloneDX S | 15 | Oct 24, 2026 | Apache-2.0 | low | 57 |
| Aegis Credential isolation proxy that injects secrets at the network boundary, with domain restrictions, agent authentication and audit logging. | 14 | Aug 6, 2026 | Apache-2.0 | moderate | 56 |
| Vuln nist MCP server Query the NIST National Vulnerability Database (NVD) API. | 14 | Sep 23, 2025 | MIT | moderate | 28 |
| Assay MCP Server Fail-closed policy-as-code proxy for MCP that denies risky tool calls, produces offline-verifiable evidence bundles and enforces egress via | 13 | Oct 23, 2026 | MIT | low | 61 |
| Koma Core MCP ◆ Classify input for prompt injection and out-of-scope content with Koma Gate's LLM-based classifier. | 13 | Oct 13, 2026 | MIT | moderate | 71 |
| Lingua Universale MCP Server Verify agent communication protocols with session types and Lean 4 proofs to catch deadlocks and role violations; includes a linter, formatt | 12 | Aug 18, 2026 | Apache-2.0 | low | 56 |
| Aletheia Runtime Safety & Scope Creep Filter Deterministic pre-execution filter for agent tool calls that blocks scope creep (credential reads, SSRF, destructive shell/SQL) and prompt i | 12 | Sep 23, 2026 | — | high | 46 |
| Authmcp gateway Auth proxy for MCP servers: OAuth2 with DCR, JWT, RBAC, rate limiting, multi-server aggregation and a monitoring dashboard. | 12 | Oct 20, 2026 | MIT | low | 56 |
| Aperion Shield Local guardrail proxy for MCP servers: blocks destructive tool calls, pins tool catalogs against rug pulls and scans for tool poisoning and | 11 | Oct 4, 2026 | — | moderate | 50 |
| Scopeblind gateway Security gateway that wraps MCP servers with per-tool policies, approval gates and optional signed receipts, in shadow (log-only) or enforce | 11 | Jul 3, 2026 | MIT | low | 48 |
| Opnsense MCP OPNsense firewall operations via API: query ARP, DHCP, firewall rules, logs, interfaces, system status and packet captures. | 11 | Oct 23, 2026 | MIT | low | 55 |
| Secretctl Secrets manager that injects credentials into commands as environment variables so agents never see plaintext secrets, and sanitizes command | 11 | Jan 22, 2026 | Apache-2.0 | low | 35 |
| Zitadel MCP Zitadel identity management: manage users, projects, OIDC apps, roles and service accounts in natural language. | 11 | Oct 21, 2026 | MIT | low | 55 |
| Agentleak Local privacy testing for AI agents: scan code for PII and secrets, score execution traces across internal channels, redact text, and track | 11 | Oct 24, 2026 | MIT | low | 55 |
| PerspectiveGraph ◆ Attack-path engine for cloud and Kubernetes: list routes from internet exposure to sensitive assets, explain each hop, find choke points and | 9 | Oct 24, 2026 | Apache-2.0 | moderate | 70 |
| Ciphertrust manager MCP server Thales CipherTrust Manager integration for key management, cryptographic operations and compliance monitoring. | 9 | Sep 5, 2026 | MIT | low | 50 |
| Hexforge gateway APK reverse engineering workspace that orchestrates jadx, apktool, adb and Frida through a workflow engine, on Termux (Android) or PC. | 9 | Oct 23, 2026 | MIT | low | 55 |
| EchelonGraph CVE & Exposure ◆ Keyless, read-only CVE and internet-exposure intelligence: CVE lookup and search, CISA KEV, EPSS history, CWE, vendor advisories (MSRC, Red | 8 | Oct 17, 2026 | MIT | low | 69 |
| OPA MCP OPA and Rego policy toolkit wrapping the OPA CLI and Regal: format, lint, evaluate, test and benchmark policies, manage the OPA REST API and | 8 | Oct 24, 2026 | MIT | moderate | 59 |
| Redacta Privacy boundary for healthcare AI agents: pseudonymises patient identifiers (NHS numbers, names, dates of birth, addresses; UK and US) into | 8 | Oct 6, 2026 | MIT-0 | low | 59 |
| Commit — Supply Chain Risk Scoring Supply chain risk scoring for npm, PyPI, Cargo and Go packages from behavioral signals: publisher depth, release consistency and maintenance | 8 | Oct 6, 2026 | MIT | low | 59 |
| Job verify Check recruiters and job offers for scams: cross-checks company registration, domain age, look-alike domains, blocklists and crypto-scam dat | 8 | Jul 3, 2026 | MIT | low | 47 |
| Bunkerweb MCP ◆ Official BunkerWeb WAF integration to manage services, instances, IP bans, configurations, plugins, and scheduler jobs over stdio or streama | 8 | Oct 23, 2026 | MIT | low | 64 |
| MCP audit Transparent proxy that intercepts, signs, rate-limits, redacts and audits MCP tool calls without modifying client or server. | 8 | Oct 18, 2026 | Apache-2.0 | low | 54 |
| Rohan protocol MCP Stateless privacy shield, WebMCP semantic firewall, and gasless on-chain state settlement for autonomous AI agents on Midnight Network. | 8 | Oct 22, 2026 | MIT | low | 54 |
| MCP bastion Security proxy for MCP servers: tool-definition pinning, tool-poisoning and exfiltration detection, injection blocking, secret redaction and | 7 | Oct 20, 2026 | Apache-2.0 | low | 58 |
| Minreestr MCP Search the Russian software registry for import-substitution and FSTEC/FSB-certified products, with full-text search and manufacturer listin | 7 | Apr 17, 2026 | MIT | low | 33 |
| Arai Enforces rules from existing agent instruction files (CLAUDE.md, .cursorrules) by blocking prohibited tool calls and logging per-rule compli | 7 | Oct 20, 2026 | Apache-2.0 | low | 53 |
| Urldna MCP URL scanning and phishing triage: capture DOM snapshots, network requests and screenshots, and hunt historical scans with a custom query lan | 7 | Aug 21, 2026 | Apache-2.0 | low | 48 |
| Lachesis Code property graph for C, Python and TypeScript: callers/callees, data and taint flow with source-to-sink witnesses, points-to and guard/si | 6 | Oct 16, 2026 | AGPL-3.0 | low | 58 |
| Guardvibe Security scanner for vibe-coded apps (Next.js, Supabase, Stripe, Prisma and more): taint analysis, auto-fix, SARIF export, pre-commit hook a | 6 | Oct 23, 2026 | Apache-2.0 | low | 58 |
| Impri ◆ Self-hostable human-in-the-loop approval inbox: humans approve, reject or edit proposed agent actions via web, mobile, Slack, Discord or Tel | 6 | Oct 23, 2026 | MIT | moderate | 68 |
| Authbox Zero-knowledge password manager and MCP credential gateway: policy-gated agent access, deterministic passwords, BIP-39 recovery and a hash-c | 6 | Oct 7, 2026 | MIT | low | 53 |
| Rad security MCP Query the RAD Security API for Kubernetes and cloud security findings, reports and runtime data. | 6 | Sep 30, 2026 | MIT | low | 53 |
| Verify MCP Offline verification of Ed25519/JCS-signed artifacts such as receipts, manifests and audit bundles. | 6 | Jul 8, 2026 | Apache-2.0 | low | 41 |
| Cmdxray Offline safety gate for agent shell commands: rates each command as danger, caution or none and explains every flag, pipe, redirect and subs | 5 | Sep 23, 2026 | MIT | low | 52 |
| Patch tuesday MCP Microsoft Patch Tuesday triage from the MSRC Security Update Guide: monthly rollups, CVE/KB lookups, supersedence chains and EPSS/CISA KEV u | 5 | Oct 14, 2026 | — | moderate | 47 |
| Kepil Accountability layer for AI agents: per-version passports, per-job mandates with spending limits, a fail-closed action gate and a hash-chain | 5 | Oct 20, 2026 | AGPL-3.0 | low | 57 |
| PCI DSS v4.0.1 Compliance Checker PCI DSS static analysis for Go payment code: finds PAN/CVV exposure, weak crypto, missing audit logs and vulnerable deps, each mapped to a P | 5 | Oct 19, 2026 | MIT | low | 57 |
| Enigma python MCP Encrypt and decrypt messages with historically accurate Enigma machine emulators from the enigmapython library. | 5 | Jul 31, 2026 | — | moderate | 37 |
| AgentAvow Signed, recomputable safety scores for MCP servers, packages and tools, with offline-verifiable attestations and a GitHub Action to gate CI | 5 | Oct 23, 2026 | — | moderate | 42 |
| Scalekit MCP server ◆ Manage Scalekit organizations, users and SSO. | 5 | Oct 16, 2026 | Apache-2.0 | low | 62 |
| Ida headless MCP Headless IDA Pro binary analysis with concurrent sessions and Il2CppDumper and Blutter metadata import for Unity and Flutter reverse enginee | 5 | Jul 14, 2026 | MIT | low | 40 |
| Claude plugin ◆ Scan and fix a domain's email authentication (SPF, DMARC, DKIM, MX), blacklist status and domain/SSL expiry, with deterministic, validated f | 2 | Oct 20, 2026 | Apache-2.0 | low | 63 |
| MINT Protocol — Universal Work Attestation ◆ MINT Protocol work attestation for agents: cryptographically attest, verify, rate and discover agent work to build portable on-chain reputat | 2 | Sep 27, 2026 | — | moderate | 53 |
| ARGUS 3 MCP Server ◆ Query ARGUS-3 and check its status; WARDEN vets third-party MCP servers with a scored firewall, tool-definition pinning and drift detection. | 2 | Oct 23, 2026 | MIT | low | 63 |
| Basedagents ◆ Agent identity registry (reputation, capability search, task marketplace, messaging) plus a local encrypted vault that leases provider keys | 2 | Oct 22, 2026 | Apache-2.0 | low | 63 |
| AIMarket MCP Gateway ◆ SSRF-hardened web gateway with web fetch, web search and Metis verification, sharing an audited security core with Metis and ARGUS. | 1 | Oct 20, 2026 | MIT | moderate | 60 |
| Kakunin MCP ◆ Agent compliance and identity: verify an agent's X.509 certificate scope, read its behavioral risk score and append to an immutable audit tr | 1 | Sep 7, 2026 | Apache-2.0 | moderate | 55 |
| Algenta MCP Server ◆ Algenta decision engine: dataset discovery, exact queries, utility models, decision memory and governed agent runs with approvals and execut | 1 | Oct 24, 2026 | Apache-2.0 | moderate | 60 |
| Mandare ◆ Accountability for AI agent fleets: verify a signed, hash-chained ledger, check budgets, kill an agent, and issue passports and mandates, al | 0 | Oct 11, 2026 | AGPL-3.0 | low | 60 |
| ~Alter Identity ◆ Personal identity profiles under a ~name handle, paid lookups that share revenue with the person found, skill-proving quests and collectives | 0 | Sep 22, 2026 | Apache-2.0 | low | 55 |
| Shieldly io MCP ◆ Analyze AWS IAM policies and CloudFormation templates with Shieldly to flag privilege-escalation paths, wildcards and over-permissive access | 0 | Jul 18, 2026 | MIT | moderate | 48 |
| Trustscoreagent ◆ Check the reputation of AI microservices and public APIs before calling them and submit ratings afterward, via an open trust registry. | 0 | Oct 21, 2026 | Apache-2.0 | low | 60 |
Official servers
Published by the service's own vendor (per the awesome-mcp-servers list): they usually track the API closely and are better maintained.
- Semgrep 17.2K ★
- SafeDep Vet MCP 1.1K ★
- DashClaw 313 ★
- Rsigma 174 ★
- Mobb vibe shield MCP 68 ★
- Studio MCP 58 ★
- MCP panther 47 ★
- Koma Core MCP 13 ★
- PerspectiveGraph 9 ★
- EchelonGraph CVE & Exposure 8 ★
- Bunkerweb MCP 8 ★
- Impri 6 ★
- Scalekit MCP server 5 ★
- Claude plugin 2 ★
- MINT Protocol — Universal Work Attestation 2 ★
Choosing and securing
Choose according to the material: source code, binaries, or documentary sources. Review dependencies, execution privileges, and support for isolated operation. Check command traceability, result formats, and documented limitations. Compare maintenance, licensing, and official or community ownership. Look for precise explanations of transmitted data and any actions that can alter analysis artifacts, rename objects, execute programs, or interact with systems outside the local analysis environment.
Repositories and binaries may contain malicious code, credentials, or details about internal weaknesses. Use an isolated environment with restricted network access and permissions. Retrieved pages may try to redirect the agent or trigger data extraction. Confirm the authorized scope before interacting with a target. Review findings carefully, since an alert or decompiled representation alone does not establish that a weakness is exploitable in practice.
Before connecting a server to an agent with access to your data, run the AI agent security check.
Most recent
- EchelonGraph CVE & Exposure created September 27, 2026
- Mandare created September 27, 2026
- Agentgate created September 17, 2026
- Kepil created September 10, 2026
- Hexforge gateway created September 6, 2026
Other categories
Learn more
Frequently asked questions
What is the best Security MCP server?
On our grid (adoption, maintenance, framework), Semgrep (95/100), World Monitor (91/100), and SafeDep Vet MCP (90/100) come first. The right choice mostly depends on the service you need and the rights you are willing to grant.
Are these servers free?
89% declare an open-source licence and are free to use. 19 need an API key: the connected service may be paid.
Does a security alert confirm that a vulnerability exists?
An alert means a rule or analysis found something requiring examination. Its significance depends on context, execution paths, and existing safeguards. Review the affected location, reproduce the reasoning within an authorized scope, and document limitations before describing the finding as a confirmed vulnerability in a report.
Should I connect a suspicious file to my usual working environment?
Use a separate environment intended for analysis. Check whether the tool only reads the file or can also execute code and invoke external programs. Restrict access to directories, credentials, and the network. Preserve the evidence you need while avoiding unnecessary distribution of the original suspicious sample.
Related articles
- Build your own MCP server in 30 minutesOctober 24, 2026
- Automate your inbox with an AI agentOctober 18, 2026
- MCP, A2A, and tool calling: agent protocols explainedOctober 17, 2026
- Build an AI agent with n8n, without codingOctober 16, 2026
Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score