Packages

A Python MCP (Model Context Protocol) server plus supporting Ableton Live remote-script components that let an AI assistant control Ableton Live (session/transport, tracks, MIDI clips/notes, devices/parameters, automation, and browser loading/import). It also includes optional integrations for text-to-speech/audio generation via ElevenLabs MCP and a TCP/UDP hybrid mode for low-latency real-time parameter control.

A small Spring AI + MCP (Model Context Protocol) server that exposes two example MCP tools for managing a simple in-memory list of courses: list all courses and fetch a course by title. It is intended to run without an HTTP web interface (STDIO transport).

Comprehensive configuration and plugin repository for Claude Code and OpenAI Codex. Bundles 16+ pre-built plugins (Azure, GitHub, Slack, Linear, MongoDB, Supabase, Playwright, etc.), custom skills, agents, hooks for auto-formatting, and MCP server integrations. Also supports alternative LLMs via proxy configurations.

Code-Index-MCP is a local-first code indexing and search MCP server for AI coding assistants. It builds a repository index (SQLite+FTS5) from local source files using tree-sitter-based parsing and supports symbol resolution and lexical search, with optional semantic search via embeddings (e.g., Voyage AI). It can watch the filesystem for updates and optionally sync/publish index baselines via GitHub artifacts. It also provides a FastAPI REST gateway and an MCP interface for assistant integration.

ContextGraph is a governed shared-memory layer for multi-agent systems. It stores agent claims with provenance/freshness/trust metadata, enforces access control at retrieval time, and compiles token-budgeted “context packs” (optionally explainable) for each agent’s permissions. It provides a Python SDK, CLI/dashboard, an HTTP REST API, and an MCP server/tool integration, with optional Anthropic Claude Memory Tool adapter support and a Neo4j-backed self-hosted beta path.

Databricks MCP Server is a Model Context Protocol (MCP) server that exposes tools for browsing Databricks workspace objects (catalogs, schemas, tables, SQL warehouses) and for executing SQL statements against a Databricks SQL warehouse, returning results to the MCP client.

Provides an MCP server that exposes Docker operations as Model Context Protocol (MCP) tools via an HTTP/SSE/WebSocket bridge. It connects to a Docker daemon (optionally over TLS) and uses MCP Mediator to generate MCP tools for Docker client service methods, including many container/image/swarm/network/volume/registry operations.

Provides an MCP server that automates Gemini (web) interactions by driving a real browser via CDP/DevTools. It spawns/controls a background browser daemon, exposes MCP tools for chat, image generation, image upload/extraction/download, session navigation, and includes a watermark-removal step for downloaded images.

GoAI is a Go library that provides a unified, type-safe interface for working with multiple LLM providers (OpenAI, Anthropic Claude, AWS Bedrock), generating vector embeddings, and performing similarity search using vector storage (including PostgreSQL vector storage). It also claims support for streaming and provides a documented Model Context Protocol (MCP) server for Go.

Provides an MCP server (Java 21+) that connects to relational databases via JDBC and exposes tools to list schemas/tables, inspect table structure, query table data, and execute custom SQL (optionally restricted to SELECT). Supports stdio (local launch) and HTTP/SSE (remote connection) modes.

MCP server for Oracle OCI Generative AI service, enabling AI agents to access OCI-hosted LLMs for text generation, summarization, and embedding generation. Wraps OCI Gen AI APIs including inference endpoints for models like Cohere and Meta Llama hosted on OCI.

Provides an MCP server (Python) that queries PubMed using NCBI advanced search, saves result files, and performs analyses such as keyword hotspot frequency, keyword trends over time, publication count over time windows, and generates comprehensive reports.

Provides an MCP server that wraps the Rope Python refactoring library and exposes project-aware refactoring tools (e.g., move/rename symbols and modules, convert modules to packages, extract methods, inline variables, and list symbols). Intended for MCP-compatible agents (e.g., Claude) to perform safe, AST-aware transformations across a Python codebase.

Provides an MCP (Model Context Protocol) stdio proxy that connects MCP clients to Ubidots, using an environment token (X_AUTH_TOKEN) and forwarding requests to an upstream MCP base URL (default https://mcp.ubidots.com/mcp).

Provides a read-only MCP server that exposes XML Documents data (via CData JDBC Driver for XML Documents) as MCP tools for listing tables/columns and running SELECT queries. Intended for local use with MCP clients like Claude Desktop via stdio.

Provides a local MCP (Model Context Protocol) server that exposes Zoho Creator data as tools to an AI desktop client. It uses the CData JDBC Driver to translate Zoho Creator into relational tables, and offers MCP tools to list tables/columns and run read-only SQL SELECT queries over stdio.

CtxVault provides a local-first semantic memory system for AI agents. Memory is organized into directory-backed 'vaults' with independent vector indexes and optional access restrictions. It supports a CLI for managing/indexing/querying vaults, a FastAPI-based HTTP REST API for CRUD/search/write operations, and an MCP server for direct agent access (list/query/write/docs).

Doctor is an agent-oriented web discovery/crawling and indexing system. It crawls web pages, chunks and embeds text (via OpenAI through LiteLLM), stores results in DuckDB with vector search, exposes a FastAPI HTTP API for fetch/search/map navigation, and provides access to these capabilities via an MCP server endpoint (/mcp) for LLM agents.

A lightweight MCP server in Go that integrates with the Tinkoff Invest API, exposing tools for searching instruments, placing market buy/sell orders, retrieving portfolio state, and getting market data (last price, order book, candles, trading status). It can run over MCP stdio and also includes an SSE HTTP server mode.

IAM Policy Autopilot is an open-source tool (CLI + MCP server) that performs deterministic static analysis of application code to generate baseline AWS IAM identity-based policies (and optionally help fix AccessDenied errors). It targets building/iterating IAM permissions for application roles using local code inspection.