Packages

Railway MCP server enabling AI agents to interact with Railway.app platform-as-a-service — managing deployments, querying project and service status, accessing deployment logs, managing environment variables, and integrating Railway's cloud deployment platform into agent-driven DevOps and deployment automation workflows.

Wazuh is the leading open-source SIEM, XDR, and security monitoring platform. Its REST API (v2, introduced in Wazuh 4.x) is a full-featured management and query interface running on the Wazuh manager node (default port 55000, HTTPS). The API covers five primary domains: security alerts (query, filter, aggregate by rule, severity, agent, time window), agent management (inventory, status, OS details, installed packages, network interfaces), rule and decoder management (read, add, enable/disable rules), compliance reporting (PCI-DSS, CIS, GDPR, HIPAA, NIST 800-53), and cluster management (node status, health, configuration). Authentication uses JWT tokens obtained by POST /security/user/authenticate with username/password (default expiry: 900 seconds / 15 minutes). Wazuh's query API supports rich filtering using q= parameter syntax (field:operator:value with AND/OR logic), enabling complex alert correlation queries. Security agents use the Wazuh API to build automated threat detection, SOC alert triage, incident investigation, compliance posture reporting, and response automation workflows without requiring direct Wazuh dashboard access.

Weather data API providing current conditions, forecasts up to 14 days, historical data, astronomy, time zone, and sports weather via a clean REST interface.

MCP server enabling AI agents to manage Xcode projects, build and test iOS/macOS apps, control iOS simulators, manage CocoaPods and Swift Package Manager dependencies, and perform archive exports.

MCP server for Amazon Bedrock — AWS's managed foundation model service giving access to Claude (Anthropic), Titan (Amazon), Llama (Meta), Mistral, and other models via a unified API. Enables AI agents to call foundation models, manage model evaluation, and leverage Bedrock's Knowledge Bases and Agents via MCP.

MCP server monitoring the Azure Updates RSS feed to provide AI agents with the latest Microsoft Azure service announcements, preview releases, general availability milestones, and deprecation notices. Enables agents to stay current on Azure platform changes without manual monitoring.

MCP server providing 43 filesystem and process management tools for AI agents — covering file operations (read, write, copy, move, delete, search), directory management, process execution, and system interaction. A comprehensive local system automation toolkit for agent-driven file and process workflows.

Buttplug.io MCP server enabling AI agents to control intimate/haptic devices supported by the Buttplug.io open-source protocol — connecting to Bluetooth LE and other supported devices, controlling vibration patterns and intensity, and integrating haptic feedback into agent-driven interactive experiences.

MCP server providing CSS capabilities and tooling to AI agents. Enables agents to query CSS properties, validate CSS syntax, generate animations and transitions, access CSS pattern libraries, and provide styling guidance — supporting AI-assisted frontend development and design system workflows.

Local-first codebase intelligence tool that uses semantic code chunking (cAST algorithm), tree-sitter parsing, and multi-hop semantic search to analyze code repositories. Extracts architecture, patterns, and institutional knowledge from codebases. Supports semantic search, regex search, and LLM-powered code research across 30+ languages.

MCP server for Coolify — the open-source Heroku/Netlify/Vercel alternative for self-hosted application deployment. Enables AI agents to deploy applications, manage services, check deployment status, view logs, manage environment variables, and control self-hosted infrastructure through Coolify's management API.

Google Search MCP server enabling AI agents to perform Google searches via Google Custom Search Engine API — executing web searches, retrieving Google search results with URLs and snippets, searching specific sites, and integrating Google's search engine into agent-driven research and information retrieval workflows.

IDA Pro MCP server written in Rust enabling AI agents to interact with IDA Pro — the industry-standard reverse engineering and binary disassembly tool — querying disassembled code, retrieving function analysis, accessing decompiled pseudocode, and integrating IDA Pro's binary analysis capabilities into agent-driven malware analysis and reverse engineering workflows.

Kubernetes MCP server enabling AI agents to interact with Kubernetes clusters — listing and querying pods, deployments, services, and namespaces, applying manifests, scaling workloads, reading logs, and integrating Kubernetes cluster management into DevOps agent workflows for infrastructure automation and observability.

MCP server built on the rquest Rust HTTP client library — a high-performance HTTP client with TLS fingerprinting that mimics real browser TLS handshakes. Enables AI agents to make HTTP requests that bypass basic bot detection by presenting realistic browser TLS fingerprints, making it more effective than standard HTTP clients for accessing sites with anti-bot measures.

MCP Tools is a Go-based CLI for interacting with MCP servers via stdio and HTTP transports, providing commands to list tools, call tools, launch an interactive shell, run a web UI, and manage MCP configurations across development editors.

NestJS module for building MCP servers that expose AI tools, resources, and prompts. Provides decorator-based API (@Tool, @Resource, @Prompt) with full NestJS dependency injection, multiple transport support (HTTP+SSE, Streamable HTTP, STDIO), Zod schema validation, per-tool authorization, and an optional built-in OAuth authorization server.

Official MiniMax MCP server (JavaScript) enabling AI agents to access MiniMax's multimodal AI capabilities — text-to-speech with voice cloning, speech-to-text, image generation, video generation, and music generation — integrating MiniMax's Chinese AI platform into agent-driven content creation workflows.

MCP server by OpenRPC maintainer shanejonas that enables AI agents to interact with OpenRPC-described JSON-RPC APIs — including Ethereum and other blockchain node APIs. Parses OpenRPC specification documents and generates dynamic MCP tools from the spec, enabling agents to call any JSON-RPC method defined in an OpenRPC document.

MCP server for Outline — the open-source team wiki and knowledge base — enabling AI agents to read and search Outline documents, create and update wiki pages, organize collections, and integrate team knowledge into agent workflows for documentation-aware assistance and knowledge management automation.