Packages

ida-multi-mcp is an MCP server/router that enables using multiple concurrently running IDA Pro instances through a single MCP endpoint. It auto-registers instances, assigns short instance_id values, routes tool calls to the correct IDA instance via local HTTP JSON-RPC, and handles instance expiration when binaries change.

octave-mcp provides an MCP server and CLI for working with the OCTAVE structured document format. It normalizes documents into a canonical form, performs schema validation with field-level error/repair “receipts,” supports controlled compression modes (loss accounting), optionally handles YAML frontmatter plus OCTAVE bodies, and can compile schema constraints into GBNF grammars for constrained generation. It offers MCP tools such as validate, write, eject (views), and compile_grammar.

Provides MCP server integration for Oracle Cloud Infrastructure services.

Provides an MCP server (stdio for local clients and HTTP for remote clients) that wraps OpenAI Sora 2 video generation APIs. Exposes tools to create videos, check job status, list jobs with pagination, download/save completed videos, remix existing videos, and delete jobs/assets.

Provides an MCP server that enables AI agents to run Textual (Python TUI) applications headlessly and interact with them via MCP tools: lifecycle management (launch/stop), UI interaction (click/hover/type/press), state inspection (snapshots, widget queries/properties, screen stack), waiting, error checking, and screenshot capture.

Calendly's REST API for managing scheduling links, event types, scheduled events, and invitees to automate appointment booking workflows.

Enterprise SAP OData to MCP Server with AI capabilities and Cloud Foundry integration

ClickHouse is a blazing-fast open-source columnar OLAP database designed for real-time analytics over massive datasets. It stores data in column-oriented format with aggressive compression, enabling queries that scan billions of rows per second. ClickHouse accepts SQL via an HTTP interface (port 8123/8443) and a native binary protocol (port 9440), with support for many input/output formats (JSON, CSV, Parquet, Arrow). ClickHouse Cloud is the managed offering with a REST Management API for programmatic cluster management. The HTTP interface supports streaming responses, progress headers, and session-based queries — making it well-suited for agents that need to run complex analytical queries.

Provides an MCP (Model Context Protocol) server that answers questions by consulting multiple LLM providers (Anthropic, Google, OpenAI) in parallel and synthesizing the results into a final response via a single tool (`cognition_wheel`).

discord-mcp is an MCP server plus CLI that lets an AI assistant manage a Discord guild/server using natural language. It provides a large set of Discord administration tools (roles, channels, members, messages, moderation, events, automod, polls, webhooks, etc.), includes an interactive setup wizard (bot creation/token validation/invite generation and client config), and offers a health check/permission audit.

Provides a Ruby/Rails integration that exposes a Rails app as a Model Context Protocol (MCP) server, letting you publish application tools (callable functions with JSON Schema input validation) and resources (URI-templated data exposure) over an HTTP/JSON-RPC style transport, with configurable authentication, rate limiting, and security protections.

Oracle Cloud Infrastructure Oracle Integration Service.

rmcp is a Python-based Model Context Protocol (MCP) server that exposes 52 statistical analysis tools (organized into categories) backed by a curated whitelist of CRAN R packages. It can be run locally (stdio for MCP clients such as Claude Desktop; optional HTTP transport for web apps) or used via a live HTTPS endpoint with interactive docs and health checks.

An MCP server that connects AI assistants directly to a live n8n instance for full workflow lifecycle management — create, read, update, delete, activate/deactivate, execute, and monitor workflows and their executions via 15 MCP tools.

A template for building production-ready MCP servers in TypeScript with Deno, providing both STDIO and HTTP transports, middleware (rate limiting/CORS/security headers/timeouts), persistent state via Deno KV, session resumability, sandboxed execution utilities, example MCP tools/resources/prompts, and CI/CD plus multiple distribution formats (JSR, native binary, DXT, Deno Deploy).

Provides an MCP server (stdio by default; optional SSE HTTP transport) for reading and writing Microsoft Excel spreadsheets (xlsx/xlsm/xltx/xltm), including reading/writing cell text and formulas, and (on Windows) capturing sheet ranges as images.

Provides a FastMCP server that exposes Kroger shopping capabilities (store lookup, product search, and cart/order operations) to AI assistants via MCP, backed by the kroger-api Python library and Kroger API OAuth2 credentials. Includes local-only cart tracking for viewing/removal/clearing and a workflow for adding items to a real Kroger cart.

Memory Bank MCP is an MCP server that provides AI coding assistants with remote read/write access to structured memory bank files, enabling persistent project-specific context across sessions for tools like Claude, Cline, Cursor, and Windsurf.

Provides an MCP server that exposes Perplexity API capabilities as MCP tools for AI assistants, enabling real-time web search and model-based Q&A/research (search, ask, research, reasoning). Supports both stdio (typical MCP clients) and an optional HTTP deployment mode.

renderdoc-mcp is an MCP (Model Context Protocol) server that uses RenderDoc’s headless Python replay API to analyze GPU frame capture files (.rdc). It exposes many MCP tools for session/capture management, event navigation, pipeline/shader inspection, resource export/extraction, pixel diagnostics, and performance/pass analysis.