Packages

OpenGuardrails MCP server enabling AI agents to apply safety guardrails — validating agent outputs for harmful content, checking responses against configurable safety policies, enforcing content restrictions, and integrating AI safety controls into agent-driven applications requiring responsible AI guardrails.

Visual web design platform with a headless CMS API that lets developers read and write CMS content, manage sites, and trigger publishing workflows programmatically.

Binance cryptocurrency exchange MCP server enabling AI agents to interact with Binance's API — querying real-time market prices, accessing order books, checking account balances, viewing trading history, and monitoring crypto market data. Can optionally support trading operations (buy/sell orders) depending on API key scope configured.

Bing China MCP server enabling AI agents to search the web using Microsoft Bing — querying Bing's search index for web results, news, and images, returning structured search results with titles, URLs, and snippets, and integrating Bing search into agent-driven research workflows as an alternative to Google Search.

The standard Python code coverage measurement tool. Coverage.py tracks which lines and branches of Python code are executed during test runs, generating HTML, XML, JSON, and terminal reports. Integrates with pytest via pytest-cov, supports branch coverage, omit patterns, and parallel execution tracking. Essential for measuring test suite completeness in Python agent codebases.

Dev.to MCP server enabling AI agents to interact with the Dev.to developer blogging platform — fetching articles, reading posts, searching content, posting new articles, managing drafts, and integrating Dev.to's developer community platform into agent-driven content creation and research workflows.

Duffel developer-first flights and travel REST API for businesses to search, book, and manage airline tickets directly through NDC (New Distribution Capability) connections to 300+ airlines, hotels, and car hire, enabling agents, OTAs, and platforms to build travel booking products with real-time inventory, ancillary services, seat selection, and post-booking management through a single modern API. Enables AI agents to manage flight search for travel booking automation, handle offer selection and seat reservation for airline checkout automation, access booking confirmation and ticket issuance for travel fulfillment automation, retrieve order management for booking modification automation, manage ancillary services for baggage and upgrade upsell automation, handle cancellation and refund for travel reversal automation, access seat map for flight seat selection automation, retrieve real-time pricing for dynamic fare automation, manage loyalty program for frequent flyer award booking automation, and integrate Duffel with travel app, corporate booking, expense management, and OTA platforms for end-to-end flight and travel booking automation.

Comprehensive TypeScript ecosystem for building production-grade applications. Effect reimagines async programming in TypeScript with typed errors, dependency injection, structured concurrency, tracing, and data pipelines — all in a single library. The Effect type `Effect<A, E, R>` encodes success type (A), error type (E), and required context (R) at the type level, making error handling and dependency management fully type-checked. Think Rust's Result type meets RxJS meets Dependency Injection, in TypeScript.

Modern, high-performance Python web framework for building APIs with automatic OpenAPI documentation, Pydantic data validation, and native async/await support. Built on Starlette for ASGI performance and Pydantic for data models. Automatically generates interactive Swagger UI and ReDoc documentation from type hints. One of the fastest Python web frameworks — comparable to Node.js and Go.

Home Assistant MCP server optimized for Xiaozhi AI assistant integration — enabling Xiaozhi and similar AI voice/chat assistants to control Home Assistant smart home devices, query entity states, trigger automations, and integrate with Chinese smart home ecosystems through Home Assistant's REST API.

Lazy MCP — a deferred/on-demand MCP server loader that loads MCP tool definitions and server connections only when needed — reducing startup overhead for agents with many configured MCP servers, deferring tool discovery until tools are actually needed, and optimizing agent initialization time in environments with large MCP server inventories.

MCP server for querying the League of Legends wiki — providing AI agents access to champion data, ability descriptions, item statistics, game mechanics, patch history, and LoL gameplay information. Enables gaming-focused AI assistants to provide accurate game data from the community wiki.

Modern async Python ASGI web framework with first-class dependency injection, type-driven OpenAPI generation, DTO support, and layered response handling — a performance-oriented FastAPI alternative.

MCP RAG server enabling AI agents to perform retrieval-augmented generation — indexing documents into a vector store, performing semantic similarity search, retrieving relevant document chunks, and providing agents with grounded, cited answers from a knowledge base of ingested documents for question-answering and information retrieval workflows.

MCP server providing SQL Server (MSSQL) database access to AI agents. Enables agents to execute T-SQL queries, inspect schema, query tables, stored procedures, and manage SQL Server databases — supporting AI-driven data analysis and database operations against Microsoft SQL Server instances.

Narsil MCP server providing AI agents with security assessment and threat analysis capabilities — enabling LLM-powered security workflows to analyze code, configurations, and systems for vulnerabilities, perform security assessments, and integrate security intelligence into agent-driven security operations pipelines.

NewsNow MCP server enabling AI agents to access aggregated news from multiple sources — fetching headlines, articles, and trending topics from news feeds and aggregators, filtering by category or keyword, and integrating current news data into agent-driven research, monitoring, and content creation workflows.

MCP server for Packmind — a platform that captures, scales, and enforces organizational technical decisions and coding standards. Enables AI agents to access and enforce your organization's documented technical patterns, architectural decisions, and coding conventions across codebases.

MCP server for SiYuan Note — a privacy-first, local-first personal knowledge management system with powerful block-based editing and bidirectional linking. Enables AI agents to read and write SiYuan notes, query the knowledge graph, search content, and integrate SiYuan PKM into AI-driven productivity workflows.

MCP server for Smart-Thinking memory system — enables AI agents to find relevant Smart-Thinking memories quickly and fetch full entries by ID. Provides semantic memory search and retrieval for AI agents, supporting persistent knowledge storage and retrieval across interactions.