Security MCP servers for scanning, threat intelligence and audits: let an AI assistant check code, dependencies and infrastructure for risks. 3,540 servers in this category.
Lets an AI client perform on-chain forensics and rug pull risk scoring for tokens across multiple blockchain networks.
Audit Cloudflare zones, detect configuration drift, and manage settings like DNS, WAF, and Email Routing through a centralized control plane.
Scan code, live URLs, and perform launch readiness checks directly within your AI client to identify security flaws and generate fixes.
$claude mcp add vibesafe-mcp -- npx -y vibesafe-mcp
Provides deterministic proof gates and a raw-byte write firewall to ensure AI agents only make authorized, verifiable changes to protected files.
Enforces security guardrails for Model Context Protocol traffic by filtering, redacting, and modifying tool calls and responses.
Scans MCP servers, configurations, and agent rule files to detect poisoned tools, hidden prompts, leaked tokens, and risky permissions.
Scans MCP server configurations, agent trust settings, and instruction files to detect security risks like tool poisoning and hidden instructions.
Provides an open format for agent execution receipts to verify that AI agent tool call results remain unchanged through SHA-256 digests.
Connects the REA reverse-engineering MCP server to DeepSeek Harness to inspect native binaries, applications, firmware, and runtime behavior.
Enables AI coding assistants to perform autonomous purple-team engineering, vulnerability validation, and dual-sided security remediation.
Audits Model Context Protocol configurations and tool manifests to detect security risks like unpinned packages, exposed credentials, and instruction injection.
Audits MCP configurations and AI agent bundles for security risks like credential leaks, SSRF, and command injection.
Audit MCP servers for tool poisoning, credential exposure, and supply-chain risks without executing any code.
Lints MCP server source code to detect prompt-injection vectors like Base64-encoded text and invisible Unicode characters.
Monitors local development servers, detects internet exposure, and provides commands to secure or stop processes, containers, and services.
Provides a collection of specialized MCP servers for developer tools, security, science, and productivity tasks.
Enforces identity, boundaries, and intent for AI agents through a governed runtime and signed, tamper-evident audit trails.
Provides a cryptographic evidence layer for AI agents to create signed, hash-chained session records for offline auditing and integrity verification.
Scans Large Language Models for security vulnerabilities and synchronizes findings with SynapseAudit.
$claude mcp add synapse-audit -- npx -y @digidenone/synapseaudit-mcp
Provides a behavioral biometric sensoric engine that turns human interaction patterns into AI-readable signal vectors and confidence scores.
$claude mcp add edna -- npx -y edna-engine