Skip to content

MCP servers for security

350 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. CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Offline review that flags agent changes which weaken tests, assertions or CI

    Runs locally
    • Open source
    • 4 Hacker News points
    • Developer tools
  3. Postgres gateway for AI agents with table grants, column masking and approval for writes

    Runs locally
    • Open source
    • 5 GitHub stars
    • Databases
  4. Self-assess AI agent governance: A3 scorecard, ISO 42001 evidence and PII scan

    Runs locally
    • 6 tools
    • Open source
    • Security
  5. Escrow for AI agent purchases on NEAR testnet, paid out only after verified delivery

    Runs locally
    • 2 tools
    • Open source
    • Finance and payments
  6. Read-only audit of agent skills in one workspace, plus a package manager for them

    Runs locally
    • Open source
    • Developer tools
  7. Experimental agent wallet that seals agreements in a public register anchored in Bitcoin

    Runs locally
    • 13 tools
    • Open source
    • Finance and payments
  8. Passive audit that checks whether an MCP endpoint exposes its tools without authentication

    Runs locally
    • Open source
    • Security
  9. Shared, audited git, test and workflow tools for MCP coding apps that never push

    Runs locally
    • 11 tools
    • Open source
    • Developer tools
  10. Read-only checks of an AI agent's Sigvara standing, score, stake and slashes

    Runs locally
    • Open source
    • Security
  11. A lockfile for MCP tools: blocks changed tool definitions and signs a receipt per call

    Runs locally
    • Open source
    • Security
  12. Gateway that verifies what an MCP server really does, then confines every call

    Runs locally
    • Open source
    • Security
  13. Scan PDF, Office, RTF and HTML files for hidden text before an agent reads them

    • Remote
    • Free trial
    • Files and documents
  14. Encrypted secrets for agents plus a phone approval step for sensitive actions

    Runs locally
    • Free trial
    • Security
  15. Scan, read and write another process's memory from an AI assistant

    Runs locally
    • 1 tool
    • Open source
    • Developer tools
  16. Gate every MCP tool call behind signed authorization and single-use capabilities

    Runs locally
    • Open source
    • Security
  17. Browser actions for AI agents that are classified, approved and logged

    Runs locally
    • 1 tool
    • Open source
    • Browser automation
  18. MCP wrapper that holds database writes until a second person approves them on their phone

    Runs locally
    • Open source
    • Databases
  19. Gives coding agents your API keys without letting the model ever read them

    Runs locally
    • 4 tools
    • Open source
    • Security
  20. Read-only Trino access for agents, run as the verified user so Trino policy decides

    Runs locally
    • 3 tools
    • Open source
    • Databases
  21. Rank security scanner findings by exposure, criticality and exploitability

    Runs locally
    • Open source
    • Security
  22. Keep API keys in your OS keychain and give coding agents only what policy allows

    Runs locally
    • Open source
    • Security
  23. Local anonymiser that swaps client names and amounts for codes before Claude reads files

    Runs locally
    • Free trial
    • Security
  24. Strip GPS, EXIF, C2PA and AI prompts from photos, video, PDFs and RAW files locally

    Runs locally
    • 2 tools
    • Open source
    • Security
  25. Local, reversible masking of client names and numbers before text goes to an AI

    Runs locally
    • Open source
    • Security
  26. Open-source MCP gateway that pins tools, holds risky calls and limits egress

    Runs locally
    • Open source
    • Security
  27. Open-source MCP server scoring pull-request risk from blast radius and git history

    Runs locally
    • 4 tools
    • Open source
    • Developer tools
  28. Policy-gated agent spending with signed receipts and audit against the payment rail

    Runs locally
    • 5 tools
    • Open source
    • Finance and payments
  29. MSP-focused MCP server and CLI for WatchGuard Cloud across customers

    Runs locally
    • Open source
    • Security
  30. Read-only payment preflight for AI agents: pay, review or do-not-pay before signing

    • 9 tools
    • Remote
    • Open source
    • Security

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