Security MCP Servers

MCP server for Gildara — operating contracts for AI agents. Structured runtime instructions compiled into optimal system prompts.

The first open-source MCP server for European cybersecurity regulations. Query DORA, NIS2, GDPR, EU AI Act, and more directly from Claude.

Security-first MCP server for 1Password — vault/item tools, prompts, resources, and op_run secret injection (MCP 2026-07-28)

MYOB Business MCP server — direct API integration with Alfe OAuth credentials and automatic token refresh

Lean Gmail MCP server with auto authentication support (maintained fork of GongRzhe/Gmail-MCP-Server)

Model Context Protocol server for StablecoinX. Auth via thirdweb email-OTP, then drive business account CRUD, API keys, payment sessions, the paymaster surface (clients, policy, activity, balance), webhooks, and floYsh yield sites and vault owner settings

California, US, and global climate compliance scans with source-linked deadlines.

Civic Auth integration for MCP servers

Self-hosted OAuth 2.1 Authorization Server for MCP servers (Express/TypeScript). Implements the MCP Authorization spec: Client ID Metadata Documents (CIMD), Dynamic Client Registration, PKCE, resource indicators, device flow.

Gmail MCP Server with on-demand authentication for SIYA/Claude Desktop. Complete Gmail integration with multi-user support and OAuth2 security.

Pre-flight MCP tool-contract drift interceptor from mcpindex.ai, the drift-monitored MCP server index. One-line wrap() over an MCP client session that HOLDs a tool call before your agent runs it when the contract drifted from your pin. No credential handl

Deterministic Scope 1/2/3 GHG inventories with bundled factors and audit hashes.

AgentSecrets MCP bridge — zero-knowledge credential tools for AI agents (Model Context Protocol, stdio)

Drop-in OAuth resource-server + connection gate for an MCP server fronted by Lane. Transport-agnostic core; adapters wrap it.

AtlaSent MCP server: with an API key, an AI agent's consequential actions wait for organizational approval and get signed proof your auditor can check. Without a key it runs an offline, unsigned demo.

MCP client plugin for bread — connect agents to MCP servers (config-level or per-agent) over stdio, Streamable HTTP, or legacy SSE, with credential-aware auth signing.

MCP proxy that checks each routed tools/call against your rules before forwarding it, and signs a record. Open source (Apache-2.0).

MCP server for using the REMnux malware analysis toolkit via AI assistants

OpenAgent middleware for the Model Context Protocol — adds OAS identity, Arsenal capability tokens, and AEGIS authorization to any MCP server in three lines.

A safe, batteries-included Model Context Protocol (MCP) server that exposes common security-testing tools and OWASP-aligned playbooks to AI clients — for AUTHORIZED testing only.

Model Context Protocol server exposing TrustRail governed-action evaluation to MCP-capable agents

Security toolkit for AI agents. Scan your machine for dangerous skills and MCP configs, monitor for supply chain attacks, test prompt injection resistance, and audit live MCP servers for tool poisoning.

CrowdStrike Falcon MCP Server

See what your AI agents can access. Scan MCP configs for exposed secrets, shadow APIs, and AI models. Generate AI-BOMs for compliance.