MCP-Saptiva is a Model Context Protocol server for interacting with Saptiva AI's models and tools through Claude Code.

• Chat (saptiva_chat): Converse with Saptiva AI models — Turbo (fast), Cortex (reasoning), Ops (RAG/web search), Legacy (tool-compatible), and KAL (Mexican legal context) — with configurable parameters like temperature, max tokens, and top-p.

• Step-by-Step Reasoning (saptiva_reason): Use Saptiva Cortex for complex math, logic, and multi-step analysis with visible chain-of-thought output.

• OCR (saptiva_ocr): Extract text from images via URL or base64 encoding, useful for document processing and receipt scanning.

• Text Embeddings (saptiva_embed): Generate semantic embedding vectors for similarity search, clustering, and RAG applications.

• Batch Embeddings (saptiva_batch_embed): Efficiently generate embeddings for multiple texts in a single call.

• List Models (saptiva_list_models): Retrieve all available models with their capabilities, descriptions, and pricing.

• Help & Docs (saptiva_help): Access guides, examples, cURL commands, and Saptiva Agents SDK information across various topics.

1. Click on "Install Server".

2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.

3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@MCP-Saptiva use Saptiva Cortex to solve this complex logic puzzle"

That's it! The server will respond to your query, and you can continue using it as needed.

Here is a step-by-step guide with screenshots.