Security MCP servers for scanning, threat intelligence and audits: let an AI assistant check code, dependencies and infrastructure for risks. 3,541 servers in this category.
Verifies signed digital receipts and audit bundles offline using Ed25519 signatures and independently provided public keys.
$claude mcp add verify -- npx -y @veritasacta/verify
Prevents AI coding agents from installing risky, deprecated, or vulnerable npm packages by enforcing dependency policies.
$claude mcp add pkgdiet -- npx -y pkgdiet
Enforces non-custodial spending policies for AI agents on the Kaspa L1 blockchain using deterministic covenants.
$claude mcp add policyvault -- npx -y policyvault-mcp
Search, audit, and install security-graded AI agent skills and MCP servers directly from your AI client.
$claude mcp add agentskillshub-mcp -- npx -y @agentskillshub/mcp
Inspects AI agents and MCP configurations for security risks, permissions, and tool access through static analysis.
$claude mcp add agentverify -- npx -y agentverify
Lets an AI agent check EVM addresses on Base and Ethereum for risks like scams, phishing, and address poisoning before sending funds.
$claude mcp add verdix-mcp -- npx -y verdix-mcpEnforce local spend caps and generate Ed25519-signed provenance receipts for AI agent tool calls and model usage.
$claude mcp add agentguard-mcp -- npx -y @agentguard-run/mcp
Enables an AI client to act as a Chief Security Officer by monitoring CVEs, scanning for secrets, vetting MCP servers, and managing incident timelines.
$claude mcp add topy-ai-cso-skills -- npx -y @topy-ai/cso-mcp
An autonomous bug bounty hunter that performs surface discovery, vulnerability testing, and evidence collection against authorized targets.
Enables AI clients to perform comprehensive pre-production security, performance, and scale audits using a massive checklist of 4,685 items.
$claude mcp add pre-production-checklist -- npx -y prodcheck
Inspect npm packages and MCP servers using local static analysis to receive security verdicts with cited evidence before installation.
$claude mcp add pkgxray -- npx -y pkgxray
Verifies AI coding assistant work by running changes in sandboxed workers and performing independent test reruns and secret scans.
$claude mcp add nomarmy -- npx -y nomarmy
A local proxy that prevents tool drift by fingerprinting MCP server tool definitions and blocking sessions if unauthorized changes are detected.
$claude mcp add mcp-pin -- npx -y mcp-pin
Provides a self-hosted gateway that issues short-lived, owner-approved API passes to AI agents to prevent them from accessing raw API keys.
$claude mcp add --transport http agentkeygate https://{host}/mcp
Creates a local, signed, and tamper-evident log of AI agent actions while providing a human-in-the-loop approval gate for consequential tasks.
$claude mcp add lotor -- npx -y lotor-mcp
Lets AI agents preflight, approve, and prove consequential actions using signed evidence and x402 payment tools.
$claude mcp add --transport http hermesplant-mcp-server https://mcp.hermesplant.com/mcp
Provides fail-closed runtime security for AI agents by intercepting tool calls to block command injection, SSRF, and credential exfiltration.
$claude mcp add ccs-mcp-server -- npx -y ccs-mcp-serverProvides fail-closed runtime interception for AI agents to block command injection, SSRF, and credential exfiltration before execution.
$claude mcp add agent-runtime-guard -- npx -y agent-runtime-guard
Enables secure repository intelligence synchronization, malware review, and dependency vulnerability scanning via source-free payloads.
$claude mcp add weavatrix-online -- npx -y weavatrix-online
Enables source-free repository intelligence synchronization, malware review, and dependency vulnerability scanning via secure cloud or self-hosted endpoints.
$claude mcp add weavatrix-online -- npx -y weavatrix-online