"Kuma" matching MCP servers:

Multi-model vision understanding MCP server that provides unified image analysis for AI assistants without native vision, supporting models like GLM-4.6V, DeepSeek-OCR, Qwen3-VL-Flash, and more.

Blocks prompt injection, jailbreaks, and out-of-scope input before an AI agent acts. classify_input tool, stdio transport.

An MCP server that adds visual understanding to text-only LLMs via image understanding, OCR, and image comparison tools, with multi-provider fallback and context-aware Focus Hint for precise descriptions.

A local RAG context optimizer MCP server that filters retrieved context to the minimum an LLM needs to answer correctly, using a cross-encoder and optional calibration.

Enables image and video generation using Luma's API, supporting multiple models, aspect ratios, and advanced features like keyframes and style references.

Enables natural-language interaction with SAP BTP destinations and OData services through MCP tools, allowing users to discover and query OData endpoints and get AI-generated answers.

A thin MCP server that delegates lightweight tasks from Claude Code or any MCP-compatible client to local or cloud LLMs via LiteLLM, supporting models like Ollama and cloud APIs as subagents.

Enables AI assistants to retrieve current weather conditions and forecasts for any location via natural language, with optional AI-generated summaries.

Enables AI assistants to query KumoRFM for predictive analytics on relational data, including graph management, natural language to PQL conversion, and training-free predictions.

Enables Claude Code to use Alibaba Cloud Token Plan models via a local proxy that maps Anthropic-style model names to Alibaba IDs and provides missing endpoints like model discovery.