Packages

Eion MCP server providing persistent agent memory via a purpose-built memory database — storing and retrieving agent memories, managing long-term context across conversations, and integrating Eion's agent-native memory system into agent workflows that require persistent knowledge and cross-session memory.

Elastic's managed cloud service API for provisioning and managing Elasticsearch and Kibana clusters, with access to the full Elasticsearch REST API for search, log analytics, and vector search.

Express REST API + MCP Server Framework combining an Express.js REST API with an MCP server in a single Node.js application — providing boilerplate for building services that expose both HTTP REST endpoints and MCP tools from the same codebase, enabling developers to build dual-interface services for AI agents and traditional API consumers.

MCP server for fast domain name availability checking and WHOIS lookups. Enables AI agents to check if domain names are available for registration, retrieve WHOIS information, query DNS records, and support domain research workflows — useful for brand naming, product launches, and domain portfolio management.

Google Cloud Functions REST API for deploying, invoking, and managing serverless HTTP and event-driven functions on Google Cloud Platform.

Google Calendar REST API for creating, reading, updating, and deleting calendar events, managing multiple calendars, and setting up event notifications.

Google Cloud Document AI REST API for intelligent document understanding, OCR, form parsing, and structured data extraction from documents. Enables AI agents to manage document OCR and text extraction for document digitization automation, handle form and table parsing for structured form data extraction, access pre-trained processors (invoice, receipt, ID, contract, W-9) for common document type extraction, retrieve entity extraction and relationship mapping for complex document understanding, manage document splitting and classification for automated document routing, handle optical character recognition for 200+ languages for multi-language document processing, access hitl (Human-in-the-Loop) workflow integration for quality-controlled extraction review, retrieve confidence scoring and bounding box data for extraction quality assessment, manage batch processing for high-volume document extraction jobs, and integrate Document AI with BigQuery, Cloud Storage, and enterprise data platforms.

Headless Gmail MCP server enabling AI agents to interact with Gmail without a browser UI — reading emails, searching inbox, sending messages, managing labels and filters, and automating email workflows using Gmail's API. 'Headless' indicates it operates without a visible browser, suitable for server-side automation and CI/CD pipelines.

MCP server for Kirby CMS — a popular file-based PHP content management system known for its flexibility and developer-friendly approach. Enables AI agents to read and manage Kirby site content, query pages, update content fields, manage files, and interact with Kirby's panel API — supporting AI-driven content management workflows.

Community MCP server for Kubernetes enabling AI agents to interact with K8s clusters — querying pod/deployment/service state, reading logs, describing resources, managing namespaces, and performing kubectl-equivalent operations.

Luma AI MCP server enabling AI agents to use Luma Labs' generative AI capabilities — generating videos with Dream Machine (text-to-video and image-to-video), creating 3D assets from images using Luma's NeRF/3DGS technology, and integrating AI video and 3D generation into agent-driven creative workflows.

MCP server that executes Python code in a sandboxed WebAssembly environment using Pyodide running in Deno. Allows AI agents to run Python snippets with dependency installation, capturing stdout/stderr and return values. ARCHIVED: Project retired January 2026 in favor of Monty.

SSH MCP server enabling AI agents to execute commands on remote servers via SSH — connecting to SSH hosts, running commands, reading output, managing files on remote systems, and integrating remote server access into agentic infrastructure workflows. Similar to mcp-ssh-manager but potentially a simpler or different implementation.

SSH Manager MCP server enabling AI agents to manage SSH connections and execute remote commands — listing configured SSH hosts, connecting to remote servers, running commands over SSH, managing SSH keys, and integrating remote server access into agentic infrastructure management workflows. Provides a safe, structured interface for SSH operations from AI agents.

HashiCorp Terraform Cloud's REST API for managing workspaces, runs, state files, variables, and modules. Enables programmatic control of infrastructure-as-code workflows and pipelines.

An MCP server that bridges Hugging Face Spaces with Claude Desktop, enabling AI agents to invoke ML services hosted on HF Spaces including image generation (FLUX, Shuttle), speech-to-text (Whisper), text-to-speech (Parler TTS), audio generation, vision models, and chat models — with automatic file handling and endpoint detection.

n8n2mcp bridge server enabling AI agents to trigger and interact with n8n workflows via MCP — exposing n8n workflow webhooks as MCP tools, allowing agents to invoke complex automation pipelines, data transformations, and integrations built in n8n from within their tool-calling context.

Type-safe settings management with Pydantic — loads configuration from environment variables, .env files, secrets files, and custom sources with automatic type coercion and validation. pydantic-settings features: BaseSettings subclass for defining settings schema, automatic env var loading, case-insensitive env var matching, nested model support (model_config SettingsConfigDict), env_prefix for namespacing, secrets_dir for Docker secrets, JSON field support in env vars, .env file loading via dotenv, custom settings sources (YAML, TOML, remote), model_validator for cross-field validation, and computed_field for derived settings.

Library that wraps existing stdio-based MCP servers into AWS Lambda functions, supporting multiple connection methods (API Gateway, Bedrock AgentCore, Lambda Function URLs, Lambda Invoke) with OAuth and IAM authentication.

The canonical Python machine learning library for traditional (non-deep-learning) ML, providing a consistent fit/predict API across classification, regression, clustering, dimensionality reduction, preprocessing, and pipeline construction.