Packages

estonia-ai-kit is an open-source monorepo that provides integration layers (MCP servers, CLI tools, and Claude Code plugin skills) to connect AI applications to Estonian digital infrastructure and services (e.g., business register/open data) and includes authenticated workflows via Smart-ID/ID-card for some services such as EMTA and LHV.

GolemCore Bot is a Java/Spring Boot framework and runnable agent that supports skill-based behaviors (Markdown SKILL.md with YAML frontmatter), MCP-based tool server integrations (stdio), tiered multi-LLM routing, optional memory and RAG, and built-in sandboxed tools (with confirmation for destructive actions). It can run as a CLI/Telgram bot and provides a web dashboard and HTTP triggers (webhooks).

Provides a Dockerized set of Model Context Protocol (MCP) security lab servers with intentionally vulnerable and corresponding hardened implementations for multiple common vulnerability classes (e.g., path traversal, SQL injection, prompt/tool-response injection, command injection). Includes challenge walkthroughs and proof artifacts meant to reproduce exploit-to-mitigation flows end-to-end.

mcp-server-aws-core is an MCP server that exposes core AWS capabilities to an MCP-capable agent, enabling the agent to perform AWS-related actions (e.g., querying resources and invoking AWS operations) using AWS credentials configured for the server.

Provides an MCP server/CLI that connects LLM agents to Telegram via Telethon (MTProto), enabling sending/editing/deleting messages, searching dialogs, managing drafts, and downloading media through MCP tools.

OpenEcon Data is an AI-assisted economic data aggregator that routes natural-language requests to multiple official data providers (e.g., FRED, World Bank, IMF, Eurostat, BIS, Comtrade, plus crypto/currency sources). It provides charts and exports, and exposes access to agents via an MCP server endpoint over SSE. It also includes a web UI (data.openecon.ai/chat).

An MCP server that exposes Oracle Cloud Infrastructure (OCI) Migration service APIs, enabling AI agents to programmatically manage cloud migrations including discovering source environments, creating migration plans, and executing workload migrations to OCI.

An MCP server that exposes Oracle Cloud Infrastructure (OCI) Search with OpenSearch service APIs, enabling AI agents to create and manage managed OpenSearch clusters, configure cluster sizing and security, manage indexes, and integrate full-text search and analytics capabilities into OCI-hosted applications.

Provides a read-only MCP (Model Context Protocol) server that wraps a CData JDBC Driver for SAP Sybase, exposing Sybase data as SQL-queryable tools (get_tables, get_columns, run_query) over stdio for use by local MCP clients (e.g., Claude Desktop).

Provides an MCP server (stdio and SSE) for interacting with the SonarQube Cloud (SonarQube Cloud API) from an MCP-capable client, using a SONAR_TOKEN for authentication.

Provides an MCP (Model Context Protocol) server that exposes tools for retrieving sports data such as team search, league tables, fixtures, event details, and player profiles across multiple sports (notably football/soccer and US sports like NBA/NFL/NHL/MLB). The README indicates it is free and does not require an API key, implying it uses an upstream free tier (e.g., TheSportsDB) for data access.

Provides a read-only Model Context Protocol (MCP) server that exposes Basecamp data through CData JDBC Drivers, allowing an MCP-capable client (e.g., Claude Desktop) to discover tables/columns and run read (SELECT) queries without writing SQL.

CodePrism is an MCP (Model Context Protocol) server that performs graph-based code intelligence and exposes multiple analysis/search/orchestration tools (e.g., repository stats, symbol/content search, dependency tracing, complexity/security/performance analysis). It can run over MCP transports such as stdio, HTTP, and SSE (per the README’s Mandrel harness description).

Provides a Model Context Protocol (MCP) server that exposes Elasticsearch capabilities as MCP tools, implemented using Quarkus. It starts an SSE endpoint and a streamable HTTP endpoint and forwards requests to a configured Elasticsearch cluster (supports optional basic username/password).

Autonomous research agent that plans research questions, scrapes and validates multiple web and local sources, then synthesizes findings into detailed cited reports exceeding 2000 words with 20+ sources.

An MCP (Model Context Protocol) server that exposes Manticore Search capabilities to MCP-compatible clients, including search, document insert/update/delete, table/index inspection, and cluster status/management tools.

A lightweight, annotation-driven Java SDK/framework for building MCP (Model Context Protocol) servers. It lets you define MCP resources, tools, and prompts via Java annotations, and run an MCP server in modes like STDIO, STREAMABLE (recommended), and SSE (deprecated) using a YAML configuration.

An MCP (Model Context Protocol) server that exposes Tekton (pipeline/task/pipelinerun/taskrun) operations as MCP tools, including listing, creating/updating/patching, deleting, starting/restarting, and integration with Artifact Hub to discover/install/trigger Tekton tasks and pipelines.

Provides an MCP (Model Context Protocol) server wrapper around PaddleOCR functionality, enabling an agent to submit images for OCR and receive extracted text/results via MCP tools.

parse-server is a Node.js/Express server that exposes a Parse-compatible backend API (e.g., for Parse-style objects, queries, file storage, and push related integrations depending on configuration). It can be used with a variety of data stores (commonly via the MongoDB adapter) to provide a REST-style Parse API.