Skip to main content
Glama
Go-Business-Inc

MCP OpenAI Images

by Go-Business-Inc
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

Generar variación desde referencias (OpenAI GPT Image)

generar_variacion

Create image variations from up to 16 reference images, supporting inpainting masks, custom sizes, and branded text overlays for consistent multi-platform visuals.

Instructions

Genera una imagen nueva a partir de 1 a 16 imágenes de referencia (endpoint de edición de OpenAI). Úsalo para derivar las piezas (Blog, LinkedIn, Instagram, YouTube, Newsletter) desde la imagen maestra del Tema y mantener consistencia visual. Acepta URLs (incluidas URLs firmadas de Pipefy), rutas locales, data URLs o base64. Con mask hace inpainting de una zona. Devuelve file_path igual que generar_imagen.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
nNoCantidad de variantes (1-10). Default 1.
maskNoMáscara para inpainting (URL/ruta/base64), se aplica a la primera referencia. PNG con alfa: zonas transparentes = se editan. Si no tiene alfa, el BLANCO se interpreta como zona a editar. Se redimensiona sola.
sizeNocuadrada (1024x1024, default) | apaisada (1536x1024, 3:2) | vertical (1024x1536, 2:3) | auto. Con gpt-image-2+ también ANCHOxALTO libre (múltiplos de 16, proporción 1:3–3:1, ej. 1280x720 o 1536x864).
modelNoModelo de OpenAI. Default: gpt-image-2. Ej: gpt-image-2, gpt-image-2.5-flare, gpt-image-1.5, gpt-image-1-mini.
promptYesPrompt en inglés. Se envía tal cual a OpenAI (incluye aquí colores/estilo de marca).
crop_toNoRecorta/redimensiona el resultado a estas dimensiones exactas (ej. 1280x720 para thumbnail de YouTube).
qualityNolow | medium (default) | high. xhigh/max solo en gpt-image-2.5-*. Recomendado high para la imagen maestra del Tema.
backgroundNoFondo. transparent requiere output_format png o webp.
overlay_textNoTexto compuesto con las fuentes reales de marca (no generado por la IA). Se aplica DESPUÉS de crop_to, así que las coordenadas son sobre la imagen final.
output_formatNoFormato del archivo final. Default png.
incluir_base64NoDefault false. Si true, agrega file_name + file_base64 (compatible con upload_file_base64 de Pipefy). Ojo: una imagen en base64 pesa millones de caracteres y puede truncarse o llenar el contexto; lo recomendado es upload_file_path con el file_path devuelto.
input_fidelityNoQué tanto preservar detalles de las referencias (gpt-image-1/1.5; no aplica a 1-mini). gpt-image-2 usa alta fidelidad automáticamente.
nombre_archivoNoPrefijo para el nombre del archivo (ej. 'tema-123-linkedin'). Default: primeras palabras del prompt.
incluir_previewNoDefault true. Devuelve una miniatura (512px) para revisar el resultado visualmente.
reference_imagesYes1-16 imágenes: URL pública/firmada, ruta local absoluta, data URL o base64. Máx. 50MB c/u.
output_compressionNoCalidad 0-100 para jpeg/webp.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.0.0

TDQS

A4.2/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

readOnlyHint=false already signals mutation; the description adds that input can be Pipefy-signed URLs, local paths, data URLs or base64, and that 'mask' performs inpainting. It also anchors the output expectation by saying it 'Devuelve file_path igual que generar_imagen.' No contradiction with annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three short, information-dense sentences: action first, use case second, input/output constraints last. No filler or repeated schema boilerplate beyond one useful summary of accepted sources.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given 16 parameters and no output schema, the description covers tool selection, input formats, inpainting capability, and the key output field. Rich parameter descriptions supply the rest; only an explicit return-shape listing and clear alternatives would make it fully complete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

All 16 parameters are documented at 100% schema coverage with defaults, enums, and constraints, so the baseline is 3. The prose repeats high-level input kinds and inpainting rather than adding new per-parameter semantics, with only the Pipefy-signed URL note being extra context.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource: 'Genera una imagen nueva a partir de 1 a 16 imágenes de referencia' and identifies the OpenAI edit endpoint. It separates the tool from 'generar_imagen' by the central role of reference images and even points to the same return contract.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly says 'Úsalo para derivar las piezas... desde la imagen maestra del Tema', giving a concrete when-to-use scenario. It does not list when-not cases or name alternatives like 'componer_imagen', so it stops short of full routing guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools