Skip to content

MCP servers for security

383 MCP servers that scan for vulnerabilities, manage secrets and investigate threats. Each one shows how to install it in Claude, Cursor and VS Code, whether it runs locally or remotely, and the tools it provides.
  1. Scan prompts, tool definitions and model output for injection, and guard tool calls

    Runs locally
    • Open source
    • AI models and LLM tooling
  2. Passive-by-default OSINT lookups for AI agents: DNS, WHOIS, IP, breaches, Shodan and more

    Runs locally
    • 16 tools
    • Open source
    • Security
  3. Local security scanner that checks AI-written code and drives a scan, fix, rescan loop

    Runs locally
    • Open source
    • Security
  4. Persistent sandbox for coding agents: run commands, edit files, manage Git and sessions

    Not checked yet
    • 1 tool
    • Remote
    • Open source
    • Developer tools
  5. Open-source MCP gate that signs an allow, deny or hold before each agent tool call

    Not checked yet
    • Remote
    • Open source
    • Security
  6. Read-only CVE triage that explains each verdict from KEV, EPSS and exploit data

    Runs locally
    • 1 tool
    • Open source
    • Security
  7. Telegram MCP server with a separate key per AI client, scoped to chosen chats

    Runs locally
    • 18 tools
    • Open source
    • Communication
  8. Signed, sealed handoffs between AI agents, with trust decided by the receiver

    Runs locally
    • Open source
    • Security
  9. Give an AI agent its own DID: prove identity offline, verify bundles, pay and message

    Not checked yet
    • 1 tool
    • Remote
    • Open source
    • Security
  10. MCP server and CLI giving coding agents version-matched docs and a dependency check

    Runs locally
    • 2 tools
    • Open source
    • Developer tools
  11. Read-only operator view of Vongola, a Rust TLS reverse proxy built on Pingora

    Runs locally
    • Open source
    • Cloud and infrastructure
  12. Rust reverse-engineering MCP server that returns evidence-backed findings on binaries

    Runs locally
    • Open source
    • Security
  13. File ownership, expiring shares and audit trail for SaaS apps, exposed to agents over MCP

    Runs locally
    • Open source
    • Files and documents
  14. Read-only preflight checks for MCP and x402 endpoints before an agent depends on them

    Runs locally
    • 2 tools
    • Open source
    • Developer tools
  15. Open-source gateway that routes AI requests by privacy policy and governs tool access

    Not checked yet
    • Remote
    • Open source
    • Knowledge and memory
  16. Give an AI assistant its own Bitwarden vault for logins, 2FA codes and passkeys

    Runs locally
    • 7 tools
    • Open source
    • Security
  17. Python library that adds permission tiers, approval gates and audit logs to MCP servers

    Runs locally
    • Open source
    • Security
  18. Open-source evidence lake that grades security controls for audits

    Runs locally
    • Open source
    • Security
  19. Email MCP server for Gmail, Outlook and IMAP that fences messages against prompt injection

    Runs locally
    • 12 tools
    • Open source
    • Communication
  20. Passive scan of a website's AI-agent readiness: robots, llms.txt, JSON-LD, MCP, security

    Runs locally
    • Open source
    • Marketing and SEO
  21. Run untrusted agent shell commands in a gVisor sandbox from any MCP client

    Runs locally
    • 1 tool
    • Open source
    • Security
  22. Scan Node.js projects for unprotected API routes and get an explained health score

    Runs locally
    • Open source
    • Developer tools
  23. Self-hosted gateway that masks PII and signs receipts for AI access to your databases

    Runs locally
    • Open source
    • Security
  24. Read-only access to honeypot observations: logins, probes and captured payloads

    Not checked yet
    • Remote
    • Contact sales
    • Security
  25. Record AI agent verdicts bound to file hashes and refuse stale ones before shipping

    Runs locally
    • Open source
    • Security
  26. Compare two logged-in identities to surface broken access control and IDOR

    Runs locally
    • Open source
    • Security
  27. Manage Pi-hole v6 DNS blocking, domains, clients, DHCP and query logs from an assistant

    Runs locally
    • 10 tools
    • Open source
    • Cloud and infrastructure
  28. Docker-run MCP server for finding public emails and subdomains of a domain

    Runs locally
    • 4 tools
    • Open source
    • Security

See every kind of AI for security: tools, agents, MCP servers and skills.Ranked: the best AI for security.