Security MCP servers for scanning, threat intelligence and audits: let an AI assistant check code, dependencies and infrastructure for risks. 3,537 servers in this category.
Lets AI agents write and tune detections, query security logs, and triage alerts in Panther.
Verifies AI system claims against provided evidence bytes and replays conformance vectors using an offline, read-only environment.
$claude mcp add agent-evidence-mcp -- uvx agent-evidence-mcp
Allows AI agents to create identities, sign and verify W3C Verifiable Credentials, and scan for leaked cryptographic keys.
$claude mcp add vouch -- uvx vouch-mcp
Checks any news domain's credibility score, category, and level so AI agents can flag unreliable sources before citing them.
$claude mcp add cred-1 -- npx -y @aloth/cred1
Exposes IDA Pro's disassembler and Hex-Rays decompiler to AI clients for binary analysis, reverse engineering, and automated code mutation.
$claude mcp add idamesh -- uvx idamesh
Provides a zero-trust security control plane for MCP agents by blocking prompt injections, PII leakage, and resource exhaustion via local guardrails.
$claude mcp add mcp-bastion -- uvx mcp-bastion-python
Enables AI agents to verify identities, manage credentials, and perform trust scoring across decentralized networks and commerce platforms.
$claude mcp add --transport http moltrust-mcp-server https://api.moltrust.ch/mcp
Provides cryptographically signed, hash-chained audit trails to ensure tamper-evident decision logging for AI agents and compliance with the EU AI Act.
$claude mcp add ai-audit-trail -- uvx ai-audit-trail
Verifies proof-bound consequential AI actions at the execution edge to ensure authorized intents are not mutated or replayed.
$claude mcp add actenon-kernel -- uvx actenon-kernel
Evaluates proposed AI agent on-chain transactions to return explainable ALLOW, WARN, or BLOCK risk decisions before execution.
$claude mcp add agentic-wallet-guardian-v3 -- uvx agentic-wallet-guardian-mcp
Provides Claude with a zero-knowledge persistent memory layer using client-side AES-256-GCM encryption and local semantic search.
$claude mcp add memoraeu-memoraeu-mcp -- uvx memoraeu-mcp
Enables AI clients to search code, packages, documentation, academic papers, and vulnerability data across multiple technical sources.
$claude mcp add codesearch -- uvx skdev-codesearch-mcp
Checks public IPs, domains, CVEs, and software versions for confirmed vulnerabilities using data from NVD, CISA KEV, and EPSS.
$claude mcp add vulnerability-exposure-mcp -- uvx vulnerability-exposure-mcp
Enables AI agents to request explicit human approval before performing high-stakes actions through a secure, auditable workflow.
$claude mcp add raposa-mcp -- uvx raposa-mcp
Screens x402 payment counterparties by performing deterministic checks on addresses, transport safety, and spending policies.
$claude mcp add frisk -- npx -y frisk-mcp
Provides end-to-end encrypted secrets management and human-in-the-loop approval for AI agents using scoped machine tokens.
$claude mcp add aisecuritywarden-client -- uvx aisecuritywarden
Surfaces external file flow and enterprise event logs to help Box admins identify potential data leakage and external sharing risks.
$claude mcp add boxadm-mcp -- uvx boxadm-mcp
Evaluate AI agent reputation and trust scores, report interaction outcomes, and issue portable trust certificates to detect Sybil attacks.
$claude mcp add agent-trust -- uvx agent-trust-mcp
Provides a local guardrail for AI agents that validates actions against YAML rules to return ACT, BLOCK, or ESCALATE decisions.
$claude mcp add abe -- uvx abe-ai
Scans Android and iOS applications for security vulnerabilities and generates pentest reports mapped to OWASP MASVS, CWE, and CVSS.
$claude mcp add gomobsec-mcp -- uvx gomobsec-mcp