"author:moton16" matching MCP servers:

Zero-dependency MCP server that equips text-only LLMs (e.g., DeepSeek) with vision by converting images to text descriptions via OpenAI-compatible vision APIs, exposing an img2text tool with multi-provider fallback.

Enables semantic and keyword search over Obsidian-style Markdown knowledge bases by registering any folder as a vault, with support for incremental indexing, raw source reads, and cache-backed embeddings over MCP stdio.

Zero-dependency MCP server that enables text-only models to "see" images via a read_image tool, which forwards the image (local path, URL, data URL, or base64) to any OpenAI-compatible vision API and returns the textual description to the main model.