Skip to main content
Glama
Antonio-byte10

mcp-google

by Antonio-byte10
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

Gerar imagem no Flow

flow_gerar_imagem

Generate images in Google Flow using a Google AI Pro account and download them locally, so you can reuse them as video frames or reference images.

Instructions

Gera imagens no Google Flow (Nano Banana) com a assinatura Google AI Pro e baixa os arquivos. Os arquivos vão SEMPRE para /app/%LOCALAPPDATA%\mcp-google/saidas/flow neste PC, venha o pedido do Claude Code ou do Claude Desktop. Leva ~20-40 s. As imagens ficam no projeto do Flow e o id delas pode ser usado como quadro inicial/final em flow_gerar_video (imagem → vídeo, bom para manter o mesmo personagem entre cenas). Se demorar mais que o tempo da chamada, volta 'em andamento': chame flow_status.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
modeloNoModelo de imagem: nano-banana-2 (padrão, equilibrado), nano-banana-pro (mais detalhe e texto legível, mais lento), nano-banana-2-lite (mais rápido). Em out/2026 os três custavam 0 créditos.nano-banana-2
promptYesDescrição da imagem em português (assunto, estilo, luz, enquadramento). Texto dentro da imagem: escreva entre aspas.
projetoNoProjeto do Flow: id, link ou nome exato (veja flow_projetos). Se omitido, usa o projeto padrão (flow_projeto_padrao; de início 'Claude MCP').
proporcaoNoSe omitida: 'vertical/reels/stories' → 9:16, 'quadrado/logo/avatar' → 1:1, senão 16:9. Para virar quadro de vídeo, use 16:9 ou 9:16 (igual ao vídeo).
quantidadeNoQuantas variações gerar (1 a 4).
personagensNoNomes de personagens do Flow (criados com flow_personagem_criar; veja flow_personagens) para aparecer com a mesma aparência. Cite o nome no prompt (ex.: 'Tatuzinho acena para a câmera'). Para manter roupa/acessórios, crie o corpo inteiro do personagem e prefira veo-3.1-fast ou quality (no Lite detalhes como chapéu podem sumir).
referenciasNoImagens de referência (até 3): id de uma imagem do projeto (flow_listar) OU caminho completo de PNG/JPG/WebP do PC. O Flow usa como base: editar uma imagem ('a mesma cena à noite'), combinar elementos, manter o estilo ou o produto.
max_creditosNoLimite opcional desta geração: só use se o usuário pedir. O teto diário do mcp-google vale sempre.
qualidade_downloadNooriginal = 1K como o Flow gerou (rápido). 2k = versão aprimorada pelo Flow (mais lenta).original

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.16.1

TDQS

A4.1/5.0
Behavior4/5

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

Annotations already declare write/open-world/non-destructive, and the description adds real context beyond them: the fixed output directory, a 20-40s latency estimate, persistence of images in the Flow project, and the async 'em andamento' fallback to flow_status. What's missing is what the call returns on success (paths? ids? credit spend), which matters for a non-read tool.

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

Conciseness4/5

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

Front-loaded with the action, then the output path, latency, and reuse/fallback notes — every sentence carries information. It is dense but slightly overpacked with parenthetical asides that could be trimmed.

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?

Covers the operational facts an agent needs for a 9-param, no-output-schema tool: where files land, how long it takes, and what to do when it times out. Return-value shape (e.g. whether the file path and image id come back in the response) is left implicit.

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?

Schema coverage is 100%, so defaults, enums and per-parameter guidance are already documented. The description adds no parameter-level detail beyond echoing the image-as-video-frame idea that is already in the proporcao field description; baseline 3 applies.

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 ("Gera imagens no Google Flow (Nano Banana)") plus the download side-effect, and names the downstream sibling flow_gerar_video it feeds into. An agent can distinguish it from flow_editar_imagem and flow_gerar_video without opening any schema.

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?

Gives concrete routing: generated image ids can be reused as start/end frames in flow_gerar_video for character continuity, and a slow run routes to flow_status. It does not explicitly contrast with flow_editar_imagem (edit existing vs generate new), so the boundary with that sibling is left implicit.

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