MCP-Saptiva
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SAPTIVA_API_KEY | Yes | Your Saptiva API key, obtainable from Saptiva Lab (https://lab.saptiva.com/) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
| prompts | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| saptiva_chatC | Send a chat completion request to Saptiva AI models. Supports multiple models including Saptiva Turbo (fast), Cortex (reasoning), Legacy (tool-compatible), and more. |
| saptiva_reasonA | Use Saptiva Cortex for complex reasoning tasks. Shows the model's chain-of-thought reasoning process along with the final answer. Best for math, logic, analysis, and multi-step problems. |
| saptiva_ocrB | Extract text from images using Saptiva OCR model. Supports both URLs and base64 encoded images. Great for document processing, receipt scanning, and image text extraction. |
| saptiva_embedB | Generate semantic embeddings for text using Saptiva Embed model. Useful for similarity search, clustering, and RAG applications. |
| saptiva_batch_embedA | Generate embeddings for multiple texts at once. More efficient than calling saptiva_embed multiple times. |
| saptiva_list_modelsB | List all available Saptiva AI models with their capabilities, descriptions, and pricing. |
| saptiva_helpA | 🎓 GUÍA PARA PRINCIPIANTES - Muestra ejemplos de peticiones y respuestas de la API de Saptiva. Temas disponibles:
|
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| code_review | Review code for bugs, improvements, and best practices |
| explain_concept | Explain a technical concept in simple terms |
| write_documentation | Generate documentation for code or APIs |
| debug_help | Help debug an error or issue |
| mexican_legal | Get help with Mexican legal and regulatory context using Saptiva KAL |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Saptiva Models | List of available Saptiva AI models and their capabilities |
| Saptiva Pricing | Pricing information for Saptiva models (per million tokens) |
TDQS
Scored across 7 tools
Each tool has a clearly distinct purpose: batch_embed and embed handle embeddings at different scales, chat and reason are for different types of model interactions, ocr is for image text extraction, list_models provides metadata, and help is for documentation. No overlap or ambiguity exists between these functions.
All tools follow a consistent 'saptiva_' prefix with descriptive snake_case names (e.g., saptiva_chat, saptiva_embed, saptiva_ocr). This pattern is uniform across all 7 tools, making them predictable and easy to identify.
With 7 tools, this server is well-scoped for an AI model service, covering core functionalities like chat, reasoning, embeddings, OCR, model listing, and help. Each tool earns its place without being overwhelming or insufficient for the domain.
The tool set covers key operations for an AI platform: chat, reasoning, embeddings (single and batch), OCR, model discovery, and help. A minor gap is the lack of tools for managing resources (e.g., deleting or updating models), but core workflows are well-supported.