Security MCP servers for scanning, threat intelligence and audits: let an AI assistant check code, dependencies and infrastructure for risks. 3,538 servers in this category.
Audits website security, checks CVE exploitation status, and provides plain-English explanations for MITRE ATT&CK techniques.
$claude mcp add security-mcp -- uvx security-mcp
Enables AI agents to request multi-party approvals for actions through a hash-chained receipt system.
$claude mcp add forgingstack-quorum -- uvx forgingstack-quorumVerifies agent mandates by checking identity, principal alignment, expiry, and allowed actions before an agent performs a task.
$claude mcp add forgingstack-mandate-check -- uvx forgingstack-mandate-check
Lets an AI client identify contract owners, proxy admins, Safes, and timelocks for any EVM address through read-only chain analysis.
$claude mcp add ownerwatch-mcp -- uvx ownerwatch-mcp
Enables AI assistants to screen e-commerce orders and user events for fraud using real-time transaction analysis and feedback loops.
$claude mcp add mcp-fraudlabspro -- uvx mcp-fraudlabspro
Allows an AI client to manage KeyCloak users, groups, sessions, and security events via the Admin REST API using a service account.
$claude mcp add keycloak-mcp -- uvx keycloak-mcp
Acts as an authorization gateway between LLM agents and OPC UA servers to enforce process-aware access policies like value ranges and interlocks.
$claude mcp add otgate-opcua -- uvx otgateActs as an authorization gateway between LLM agents and OPC UA servers to enforce process-aware access policies like value ranges and interlocks.
$claude mcp add otgate -- uvx otgate
Provides a public transparency log to track, diff, and verify the tool definitions of MCP servers to prevent prompt injection via definition changes.
$claude mcp add histor -- uvx aimarket-histor
Lets an AI agent perform passive security audits on untrusted code using a hardened container of ten security scanners.
$claude mcp add appsec-toolbox-mcp -- docker run -i --rm ghcr.io/ai-ronin-systems/appsec-toolbox-mcp:0.3.0
Anchors file SHA-256 fingerprints to the Bitcoin blockchain without the files ever leaving the user's device.
$https://github.com/Orphograph/Orphograph/releases/download/mcp-v0.1.6/orphograph-mcp-0.1.6.mcpb
Redacts European personal data like names, national IDs, and IBANs from text and files to protect privacy when using LLMs.
$claude mcp add eu-pii-redact -- uvx eu-pii-redact
Acts as a deterministic gateway between LLM agents and MCP tools to verify critical arguments against trusted sources and security policies.
$claude mcp add intentplane -- uvx intentplane-mcp
Lets an AI assistant read EyalSec security events, machines, runs, and rules to help triage security findings through the EyalSec API.
$claude mcp add eyalsec-mcp -- uvx eyalsec-mcp
Provides a tamper-evident, hash-chained logbook for AI agents to record actions, tool calls, and human approvals on your own server.
$claude mcp add sovereign-agent-logbook -- uvx sovereign-agent-logbook
Lets an AI client perform experimental reverse CAPTCHA challenges to receive signed badges or credentials based on machine-oriented task performance.
$claude mcp add mettle -- uvx mettle-verifier
Provides audited, governed operations for Keycloak and authentik identity providers, enabling AI agents to manage users, clients, and security audits.
$claude mcp add identity-aiops -- uvx identity-aiopsProvides tamper-evident, hash-chained action logging for AI agents using a single JSONL file with zero dependencies.
$claude mcp add ledger -- uvx arcaeon-ledger
Lets an AI assistant manage MikroTik RouterOS devices by running commands for VLANs, firewall rules, DNS settings, and device inventory.
$claude mcp add mikrotik-mcp -- uvx mcp-server-mikrotik
Lets an AI client automate reverse engineering tasks like disassembly, debugging, Android analysis, and exploit development.