~/hackyfeed $

MCP Router is a desktop application designed for streamlined management of Model Context Protocol (MCP) servers, allowing users to connect to both local and remote servers with support for various connection methods. Key features include centralized server management with project and workspace organization, cross-platform compatibility, and robust privacy measures ensuring that all data remains local to the user's device. The tool also offers easy integration with popular AI tools and detailed logging and analytics for monitoring server activity.

MCP Proxy is a TypeScript-based HTTP and Server-Sent Events (SSE) proxy designed for MCP servers that utilize stdio transport, facilitating compatibility between clients of different protocol revisions. It enables seamless handling of protocol revisions from 2025 to 2026-07-28 using a single endpoint, with built-in CORS support, while allowing configurable connection settings for upstream negotiation. Notable features include progress tracking, cancellation support, and a streamlined architecture that allows for multi-client interactions without requiring them to be aware of each other's protocol versions.

MCP Manager is a visual client tool designed for managing Model Context Protocol (MCP) environments, providing users with an intuitive graphical interface to simplify operations that typically require complex command-line commands. Key features include one-click environment installation and uninstallation, as well as convenient management of MCP services, enabling complete lifecycle management in a user-friendly manner. Currently supported on macOS, with a Windows version under development, it caters to personal, educational, and non-commercial uses under a modified GPL-3.0 license.

The BigQuery MCP Server facilitates secure, read-only communication between AI language models and BigQuery databases, allowing users to query their data using natural language without writing SQL. Key features include auto-discovery of sensitive data fields, configurable access restrictions to prevent exposure of PII and PHI, and a validation system that ensures only non-mutative `SELECT` statements are executed. It offers two operational modes to cater to varying levels of data sensitivity, making it versatile for different use cases.

The Cloudflare MCP Server is a token-efficient server designed to interact with the Cloudflare API, offering access to around 2,500 endpoints within a compact token limit of approximately 1,000 tokens. Its primary use case is to facilitate API call execution with minimal token overhead, leveraging a unique Code Mode to dynamically generate tool configurations based on the specified API endpoints. Notable features include support for OAuth and API tokens, customizable endpoint settings, and the ability to disable Code Mode for direct API call handling, albeit at a higher token cost.

Mastra is a framework designed for developing AI-powered applications using a modern TypeScript stack, facilitating the transition from prototypes to production-ready solutions. Notable features include model routing for seamless integration with multiple AI providers, automation through agents that perform complex tasks, a graph-based workflow engine for managing multi-step processes, and robust context management capabilities. Additionally, it provides essential production tools for performance evaluation and observability, making it ideal for building and scaling reliable AI products.

The rainmana/magic-json-to-csv-converter is a Spark template designed to facilitate the development of applications that convert JSON data into CSV format. Its primary use case is to provide a streamlined environment for developers looking to quickly build and test data transformation workflows using Apache Spark. Notable features include a minimal setup ideal for local development and ease of scalability for more complex projects.

Mac Monitor is an advanced, standalone system monitoring tool specifically designed for macOS environments, facilitating security research, malware triage, and system troubleshooting. Utilizing Apple's Endpoint Security APIs, it captures and enriches a variety of system events including process, memory, and file events, while providing a user-friendly interface for real-time analysis and filtering of relevant data. Notable features include high-fidelity event enrichment, dynamic event subscription management, and advanced path muting capabilities, enhancing the tool's effectiveness in detecting and contextualizing macOS threats.

LosslessCut is a cross-platform graphical user interface for FFmpeg that enables fast and lossless editing of video and audio files. Its primary use case is for trimming and cutting large media files without quality loss, making it ideal for quickly extracting desirable segments while discarding excess data. Notable features include smart cuts, lossless merging, multi-track editing, and comprehensive metadata management, all streamlined for efficient workflows.

`lmstudio-js` is the official JavaScript client SDK from LM Studio, enabling developers to utilize local large language models (LLMs) in TypeScript applications. It provides capabilities for chat responses, text completions, and model management (loading, configuring, and unloading), alongside features for creating autonomous agents. The SDK is designed for ease of use in both browser and Node-compatible environments, offering advanced functionalities not present in typical cloud-based SDKs.

The llms.txt Generator is a tool designed to create consolidated text files from web content specifically for large language model (LLM) training and inference. It features a web interface and an API for generating both standard and full LLM text files without requiring an API key for basic usage, although it is being deprecated in favor of other endpoints. The tool utilizes web crawling and advanced text processing to compile information efficiently, though users should anticipate processing delays due to the nature of crawling and LLM operations.

LiteMCP is a TypeScript framework designed for creating Model Context Protocol (MCP) servers, facilitating the development of server-side applications that expose executable functions, resources, and reusable prompt templates. Notable features include built-in logging, error handling, and a CLI for testing and debugging, as well as support for Server-Sent Events (SSE). However, the project has been deprecated in favor of the official SDK, which offers similar capabilities.

The Linear Skill for Claude Code is a tool designed to facilitate the management of Linear issues, projects, and teams by leveraging the GraphQL API. Key features include esbuild pre-compilation for faster CLI startup, a label taxonomy system for effective categorization, automated setup checks, and capabilities for creating and managing parent-child issue relationships. Additionally, the tool supports bulk synchronization and image uploads to enhance project management processes.

LazyCodex is an advanced agent harness designed to facilitate complex codebase management within Codex, providing features for project memory, planning, execution, and verified task completion. It streamlines installation through a single command using `npx`, supports autonomous operation, and includes commands for monitoring and maintaining operational health. Notable functionalities include the ability to perform iterative tasks with Oracle verification and easy integration with Codex's marketplace for enhanced interoperability.

Langgraph is a JavaScript library designed for building and manipulating directed graphs, primarily used in applications that require the representation of complex relationships between data points. Its notable features include a flexible API for graph operations, support for various graph traversal algorithms, and integration capabilities with other JavaScript frameworks. This tool is optimized for developers looking to implement graph data structures in their web applications.

LangGraph-GUI is a graphical interface designed for seamless interaction with SvelteFlow frontend and FastAPI backend using large language models (LLMs) such as Ollama or other API services. It primarily serves to simplify the deployment and management of LLM applications by leveraging Docker for containerization and offering an intuitive user experience. Notable features include Docker Compose setup for easy environment configuration and support for both Linux and Windows users.

Langfuse is an open-source platform designed for LLM engineering that enables teams to collaboratively develop, monitor, evaluate, and debug AI applications. Notably, it offers quick self-hosting options and robust features for real-time monitoring and evaluation of AI models, making it a versatile tool for teams working on AI-driven solutions.

Label Studio is an open-source data labeling tool designed to facilitate the annotation of various data types, including audio, text, images, videos, and time series data. Its primary use case is to prepare and enhance datasets for machine learning, featuring an intuitive user interface, customizable labeling templates, and the ability to integrate with existing tools and workflows. Key features include support for exporting labeled data in multiple formats and deployment options via Docker or cloud instances.

The PocketBase JavaScript SDK provides an interface for developers to interact with the PocketBase API using both browser and Node.js environments. Its primary use case is to simplify authentication and data manipulation tasks, such as querying and managing collections within a PocketBase application. Notable features include TypeScript support, customizable authentication stores, auto-cancellation of requests, and built-in security mechanisms to prevent injection attacks in filter queries.

Joplin is an open-source note-taking and to-do application that allows users to manage notes organized into notebooks with capabilities for searching, tagging, and markdown formatting. Notable features include offline access, end-to-end encryption for secure synchronization with various cloud services, and an import function for Evernote notes. The application is cross-platform, supporting Windows, Linux, macOS, Android, and iOS, and includes a Web Clipper for saving content from browsers.

Joern-mcp is a middleware tool that facilitates AI-driven code security analysis by interfacing between the Claude Code environment and a local Joern instance. It allows users to import codebases, perform vulnerability scans, conduct taint analysis, and execute arbitrary CPGQL queries on the Code Property Graph (CPG). Notable features include advanced workspace management tools, security analysis capabilities, and easy integration with the Claude Code skill set for streamlined workflows.

Jan is an open-source alternative to ChatGPT that allows users to download and run various large language models (LLMs) locally, ensuring full control and privacy. Notable features include integration with cloud AI models, the ability to create custom assistants, and an OpenAI-compatible API for seamless interoperability with other applications. The tool prioritizes user privacy by enabling local execution of models while also supporting multiple operating systems for easy installation.

The IOS-DFU-Library is an iOS framework designed to facilitate Over-The-Air Device Firmware Update (OTA-DFU) for Nordic Semiconductor's nRF5x series chips, allowing users to easily flash new firmware onto compatible devices using an iPhone or iPad. Key features include seamless integration into iOS applications, support for varying bootloader/DFU versions, and customizable configuration parameters for firmware uploading. This library enhances device management flexibility, enabling manufacturers to address bugs and deploy new features remotely.

iChatGPT is an OpenAI ChatGPT application designed for iOS, iPadOS, and macOS, providing users with a streamlined chat interface to interact with AI models like gpt-3.5 and gpt-4. Notable features include support for historical conversations, extended timeout settings to prevent request failures, and the ability to manage multiple independent chat sessions across devices, enhancing user experience with options for text copying, voice input, and customization.

Husher is a library designed to analyze and manipulate text containing hidden Unicode characters that are imperceptible to the human eye but can be interpreted by language models. Its primary functions include encoding input strings into hidden text using zero-width characters, reversing the encoding process, and sanitizing text to remove hidden elements, all while handling various character encodings, including flag emojis and UTF-8. Notable features include the ability to seamlessly integrate with Node.js via NPX or package managers, and an interactive online demonstration tool for real-time hushing and unhushing operations.