Packages

Let MCP clients like Claude and Cursor control [Intruder](https://www.intruder.io/). For more information and sample use cases, please see [our blog post](https://www.intruder.io/blog/claude-intruder-mcp#intruder-mcp-use-cases).

MCP server for interacting with IP Find API

MCP Server for RestCSV

Provides a language-friendly distribution of a single curated JSON database (mcp-clients.json) listing MCP client names, metadata, and capability flags compatible with MCP client capability schemas. Intended for MCP servers to decide how to present resources/tools/prompts/instructions based on known client feature support.

Provides an MCP (Model Context Protocol) server for OCR on images using Tesseract OCR and GoCV/OpenCV, including configurable and auto image preprocessing (quality analysis, denoising, thresholding, deskew, contrast/brightness adjustments), worker pooling, and result caching keyed by SHA256.

Provides a Model Context Protocol (MCP) server that gives tools/agents access to a VictoriaLogs instance for read-only log exploration and querying (including stream/field discovery, stats, and embedded documentation search). Supports multiple MCP transports: stdio (default), SSE, and Streamable HTTP (with /mcp endpoint) and exposes Prometheus /metrics and health checks.

shodan-mcp is an MCP (Model Context Protocol) server that exposes Shodan capabilities as 20 callable tools for AI agents. It supports passive reconnaissance and vulnerability intelligence (CVE/CPE) and provides 4 tools that work without a Shodan API key.

[](https://smithery.ai/server/mcp-server-perplexity)

UnrealGenAISupport is a UE5.1+ plugin that provides direct in-engine access to LLM APIs (OpenAI, Anthropic, DeepSeek, Gemini, Grok, and local Ollama) plus an MCP server for AI-driven Blueprint generation, scene control, and Python/console command execution from Claude Desktop or Cursor.

This extension integrates [Framelink Figma MCP Server](https://github.com/GLips/Figma-Context-MCP)

archmcp is a local MCP (Model Context Protocol) server that scans repositories and produces compact architectural snapshots (modules/symbols/dependencies/routes/architectural patterns). It runs as a stdio-based MCP server and can also generate one-shot snapshots via a CLI.

Official collection of MCP servers from AWS Labs providing AI agents access to AWS services. Includes servers for AWS documentation lookup, infrastructure-as-code (CloudFormation, CDK, Terraform), container orchestration (EKS, ECS), serverless (SAM CLI, Lambda), cloud resource management (Cloud Control API), and AWS Support. Enables foundation models to read AWS docs, manage infrastructure, and automate cloud operations.

Provides a local Model Context Protocol (MCP) server exposing JSON-RPC tools to read, write/update, query (JSONPath), validate, and get help about JSON files—primarily for use by LLM/MCP clients via stdin/stdout.

A Rust implementation of the Model Context Protocol (MCP) that can run over HTTP (including SSE streaming) and over STDIO, exposing MCP JSON-RPC methods for resources, tools, prompts, completion, and protocol/session management. Includes configurable authentication/authorization options and TOML-based configuration/validation.

MCP server providing Oracle Cloud Infrastructure (OCI) scanning capabilities for vulnerability detection, security assessment, and compliance checking of OCI resources

MCP server that provides programmatic access to Oracle Cloud Infrastructure Threat Intelligence services, enabling agents to query threat data, indicators of compromise, and security intelligence from OCI's threat intelligence platform.

A read-only Python MCP server that exposes TA-Lib functionality via MCP tools for listing available indicators/categories, fetching indicator metadata, computing indicator values from numeric OHLCV-like inputs, and reporting version information. If TA-Lib is not installed, it still starts and returns actionable errors for TA-Lib-related tools.

yt-mcp-server is a Python MCP server that exposes YouTube-related tools: searching for videos, fetching video and channel metadata, retrieving transcripts (with timestamps), searching within a single transcript, and searching across all videos of a channel for passages matching a query. It can run as a zero-config MCP server (e.g., via uvx) and also supports running a standalone HTTP server using a specified transport.

A Python SDK for building MCP servers with progressive tool disclosure. Instead of exposing all tools at once, it organizes them into workflow stages and reveals only contextually relevant tools at each step. Supports semantic search mode that collapses 100+ tools into two meta-tools (search_tools and call_tool).