20946
Indexed

Connect the world with MCP

Find awesome MCP Servers. Build AI Agents quickly.

SiliconFlow: Ready-to-use model APIs for MCP tools — plus $1 bonus for you!

Latest MCP Servers

View All
Book_finder

19 minutes ago
Mcp Guardian
@rudraneel93

__MCP Guardian__ is a runtime security, cost, and health governance proxy that sits between AI clients (Cline, Claude, etc.) and MCP servers. It enforces YAML-configurable security policies — blocklists, allowlists, regex patterns, rate limits, and token budgets — in audit, warn, or blocking mode. __Three-layer detection engine__ combats threats at every level: regex pattern matching catches known attack signatures and 50+ secret types with Shannon entropy scoring; Zod schema validation detects malformed payloads and tool parameter tampering; an optional LLM semantic layer (Anthropic) flags prompt injection, social engineering, and jailbreak attempts that evade static rules. A multi-stage payload normalizer defeats URL/hex/unicode/HTML entity encoding bypasses before any detector runs. __Real cost tracking__ intercepts actual `tools/call` traffic, counts tokens via `tiktoken`, and surfaces per-server and per-tool spend in a live web dashboard alongside Prometheus metrics, circuit breaker status, and health probe results. __Enterprise-ready__ features include OAuth 2.1/OIDC with JWT algorithm pinning and RBAC, mTLS zero-trust networking, Redis-backed HA rate limiting, PostgreSQL audit logs, OpenTelemetry tracing, Discord/Slack alerts, a Helm chart with NetworkPolicy, and a formal STRIDE threat model. 168 tests across 16 suites (unit, fuzz, integration, E2E). Runs anywhere Node.js 18+ is available. __Install:__ `npm install -g @mcp-guardian/server`

2 hours ago
Mikhai56

6 hours ago

Featured MCP Servers

View All
Tavily Mcp
@tavily-ai

JavaScript
a year ago

Featured MCP Clients

View All

Latest MCP Clients

View All
ENTRANS

a day ago
SOCIAL-API.AI

11 days ago
SILVIA-MCP

12 days ago
ENCONVERT-MCP

12 days ago
@ENCONVERT/MCP

12 days ago
@ENCONVERT/MCP

12 days ago
PDFCHECK-MCP

14 days ago
ONTHEIA

21 days ago
STATIONONE

22 days ago

FAQ

Frequently Asked Questions about MCP Server

1

What is MCP (Model Context Protocol)?

MCP is an open-source protocol developed by Anthropic that enables AI systems like Claude to securely connect with various data sources. It provides a universal standard for AI assistants to access external data, tools, and prompts through a client-server architecture.

2

What is MCP Server?

MCP Server is a system that provides context, tools, and prompts to AI clients. It can expose data sources like files, documents, databases, and API integrations, allowing AI assistants to access real-time information in a secure way.

3

How do MCP Server work?

MCP Server work through a simple client-server architecture. They expose data and tools through a standardized protocol, maintaining secure 1:1 connections with clients inside host applications like Claude Desktop.

4

What can MCP Server provide?

MCP Server can share resources (files, docs, data), expose tools (API integrations, actions), and provide prompts (templated interactions). They control their own resources and maintain clear system boundaries for security.

5

How does Claude use MCP?

Claude can connect to MCP server to access external data sources and tools, enhancing its capabilities with real-time information. Currently, this works with local MCP servers, with enterprise remote server support coming soon.

6

Is MCP Server secure?

Yes, security is built into the MCP protocol. Server controls its own resources, there's no need to share API keys with LLM providers, and the system maintains clear boundaries. Each server manages its own authentication and access control.

7

What is mcp.so?

mcp.so is a community-driven platform that collects and organizes third-party MCP Servers. It serves as a central directory where users can discover, share, and learn about various MCP Servers available for AI applications.

8

How can I submit my MCP Server to mcp.so?

You can submit your MCP Server by creating a new issue in our GitHub repository. Click the 'Submit' button in the navigation bar or visit our GitHub issues page directly. Please provide details about your server including its name, description, features, and connection information.