Packages

BookStack MCP server enabling AI agents to interact with BookStack — the popular open-source, self-hosted wiki and documentation platform. Enables reading, searching, and creating books, chapters, and pages in BookStack; integrating BookStack knowledge bases into AI agent workflows; and making organizational documentation accessible to AI assistants.

Atlassian Confluence REST API for reading and writing pages, spaces, and content in team wikis, enabling agents to search, retrieve, create, and update documentation.

MCP server for real-time cryptocurrency price data using CoinGecko API. Enables AI agents to query current and historical prices for Bitcoin, Ethereum, and thousands of altcoins, track market cap rankings, monitor price changes, and access crypto market data for trading analysis and portfolio tracking workflows.

DataScript MCP server enabling AI agents to use DataScript — the immutable in-memory database with Datalog query language inspired by Datomic — querying structured data with Datalog, managing entity-attribute-value data models, and integrating graph-capable in-memory data storage into agent-driven data analysis workflows.

SQLite toolkit for Swift — covers the full spectrum from raw SQL to Swift-idiomatic ORM. GRDB provides: DatabaseQueue (serial SQLite access), DatabasePool (concurrent readers/serial writer), record types (FetchableRecord, PersistableRecord, TableRecord), migration support, query builder DSL, Combine publishers for reactive observation, and Swift async/await API. Alternative to Core Data and SQLite.swift for iOS/macOS. GRDB prioritizes SQLite's full power — write real SQL or use Swift query builder. DatabaseObservation (ValueObservation) notifies on database changes. Used widely in iOS agent apps requiring offline storage.

Kindly Web Search MCP server providing AI agents with web search capabilities — querying search engines (DuckDuckGo or Bing) and returning relevant search results with snippets and URLs, enabling agents to access current web information beyond their training data.

An MCP server that bridges AI models to Windows debugging via WinDbg/CDB, enabling natural language crash dump analysis, live remote debugging, and batch dump processing through a Python wrapper around the CDB debugger.

Nextcloud MCP server enabling AI agents to interact with Nextcloud — the self-hosted file storage and collaboration platform — reading and writing files, managing folders, sharing documents, accessing calendar and contacts, and integrating Nextcloud's private cloud capabilities into agent-driven file management and collaboration workflows.

Enhanced MCP server for Obsidian vault interaction via the Obsidian Local REST API plugin. Extends standard Obsidian MCP with additional capabilities: reading, creating, updating notes; managing frontmatter/properties; searching vault content; handling attachments; and managing daily notes. Provides comprehensive AI agent access to Obsidian knowledge bases.

Oracle Database MCP server enabling AI agents to interact with Oracle Database — executing SQL and PL/SQL queries, inspecting schema objects, querying data dictionary views, running stored procedures, and integrating Oracle's enterprise relational database capabilities into agent-driven data analysis and database management workflows.

MCP server for Emacs org-mode — the powerful plain-text organizational system embedded in Emacs. Enables AI agents to read, write, and manage org-mode files including TODO items, headings, tags, deadlines, scheduled tasks, and org-mode outlines. Bridges AI capabilities with the org-mode productivity ecosystem.

Data validation and settings management library for Python using type annotations. Define models as Python classes with type hints — Pydantic validates, coerces, and serializes data automatically. Core dependency of FastAPI, LangChain, and most modern Python AI frameworks. v2 is a complete Rust-backed rewrite with 5-50x performance improvements over v1.

The dominant BDD (Behavior-Driven Development) test framework for Ruby. RSpec provides a rich DSL for writing readable, expressive tests — describe/context/it blocks, expect().to matchers, subject/let helpers, shared examples, and a comprehensive mocking/stubbing system (rspec-mocks). rspec-rails integrates with Rails for controller, model, request, and system specs. Ubiquitous in Ruby/Rails projects.

Official MCP server from ScanMalware (scanmalware org) for their URL scanning and malware detection service. Enables AI agents to submit URLs for malware analysis, retrieve scan results, and access threat intelligence data — integrating automated security scanning into agent workflows.

MCP server providing serverless RAG (Retrieval Augmented Generation) capabilities from Sionic AI. Enables AI agents to perform semantic search and retrieval over knowledge bases without managing vector database infrastructure — Sionic AI handles the embedding, indexing, and retrieval pipeline as a managed service.

Durable workflow execution platform that ensures code runs to completion despite failures, timeouts, or restarts. Workflows are expressed as code (Python/Go/Java/TypeScript) with automatic state persistence. Temporal Cloud offers a fully managed service with REST API.

Python retry library that wraps functions with configurable retry logic including exponential backoff, jitter, stop conditions, and custom wait strategies via decorators or context managers.

Testing utilities that encourage testing UI components from the user's perspective rather than implementation details. The core principle: query DOM elements by role, label, text (like a user would) rather than by CSS class or component internals. Includes React Testing Library (@testing-library/react), user-event for realistic user interactions, and screen queries. The standard approach for testing React components with Jest or Vitest.

Open-source, self-hostable CRM with a GraphQL and REST API for managing contacts, companies, opportunities, and custom objects with a Notion-like flexible data model.

MCP server for the Zig programming language, from openSVM organization. Enables AI agents to interact with Zig toolchain — compiling Zig code, running Zig programs, querying Zig documentation, and supporting AI-assisted systems programming in Zig.