mcp-alphabanana
mcp-alphabanana
Englisch | 日本語
mcp-alphabanana ist ein Model Context Protocol (MCP) Server zur Generierung von Bild-Assets mit Google Gemini. Er wurde für MCP-kompatible Clients und Agenten-Workflows entwickelt, die eine schnelle Bildgenerierung, transparente Ausgaben, Referenzbild-Steuerung und flexible Bereitstellungsformate benötigen.
Schlüsselwörter: MCP-Server, Model Context Protocol, Gemini AI, Bildgenerierung, FastMCP
Hauptfunktionen:
Ultraschnelle Gemini-Bildgenerierung über Flash- und Pro-Stufen hinweg
Transparente PNG/WebP-Asset-Ausgabe für Web- und Spiele-Pipelines
Multi-Bild-Stilsteuerung mit lokalen Referenzbilddateien
Flexible Datei-, Base64- oder kombinierte Ausgaben für Agenten-Workflows

Schnellstart
Starten Sie den MCP-Server mit npx:
npx -y @tasopen/mcp-alphabananaOder fügen Sie ihn Ihrer MCP-Konfiguration hinzu:
{
"mcp": {
"servers": {
"alphabanana": {
"command": "npx",
"args": ["-y", "@tasopen/mcp-alphabanana"],
"env": {
"GEMINI_API_KEY": "${env:GEMINI_API_KEY}"
}
}
}
}
}Setzen Sie GEMINI_API_KEY, bevor Sie den Server starten.
Für Claude Desktop:
Laden Sie mcp-alphabanana-latest.mcpb herunter und fügen Sie es dann als Erweiterung in den Claude Desktop-Einstellungen hinzu. Für Windows wird empfohlen, die 'FileSystem'-Erweiterung für eine bessere lokale Dateiverwaltung hinzuzufügen.
Related MCP server: nano-banana-claude
Claude Registry
Die Claude-Registry / MCPB-Paket-Metadaten sind in manifest.json definiert und werden mit dem statischen 512x512-Icon unter images/mcp-alphabanana.png ausgeliefert.
Native sharp-Laufzeitpakete sind als optionale Abhängigkeiten deklariert, sodass .mcpb-Installationen die korrekte vorgefertigte Binärdatei auf jeder unterstützten Plattform auflösen können, ohne auf Postinstall-Hooks angewiesen zu sein.
Stabiler MCPB-URL:
https://github.com/tasopen/mcp-alphabanana/releases/latest/download/mcp-alphabanana-latest.mcpbVersioniertes MCPB-URL-Muster:
https://github.com/tasopen/mcp-alphabanana/releases/download/vVERSION/mcp-alphabanana-VERSION.mcpbSupport: GitHub Issues
MCP-Server
Dieses Repository stellt einen MCP-Server bereit, der es KI-Agenten ermöglicht, Bilder mit Google Gemini zu generieren.
Er kann mit MCP-kompatiblen Clients verwendet werden, wie zum Beispiel:
Claude Desktop
VS Code MCP
Cursor
Erstellt mit FastMCP 3 für eine vereinfachte Codebasis und flexible Ausgabeoptionen.
Glama MCP Server Badge:\
Verfügbare Tools
generate_image
Generiert Bilder mit Google Gemini mit optionaler Transparenz, lokalen Referenzbildern, Grounding und Reasoning-Metadaten.
Für Claude Desktop bevorzugen Sie outputType=file für mittlere oder große Bilder. base64- und combine-Antworten verbrauchen Claude-Kontext und können das Größenlimit des Clients erreichen. Verwenden Sie unter Windows die FileSystem-Erweiterung, um einen beschreibbaren absoluten outputPath und beliebige lokale referenceImages-Pfade auszuwählen.
Hauptparameter:
prompt(string): Beschreibung des zu generierenden Bildesmodel:Flash3.1,Flash2.5,Pro3,flash,prooutputWidthundoutputHeight: gewünschte endgültige Bildgröße in Pixeln im normalen Modusnoresize+aspectRatio+output_resolution: gibt die native Gemini-Größe ohne Größenänderung zurückoutput_resolution:0.5K,1K,2K,4Koutput_format:png,jpg,webpoutputType:file,base64,combineoutputPath: erforderlich, wennoutputTypefileodercombineisttransparent: aktiviert transparente PNG/WebP-NachbearbeitungreferenceImages: optionales Array lokaler Referenzbilddateiengrounding_typeundthinking_mode: erweiterte Gemini 3.1-Steuerungen
Modellauswahl
Eingabe-Modell-ID | Internes Modell-ID | Beschreibung |
|
| Ultraschnell, unterstützt Thinking/Grounding. |
|
| Legacy Flash. Hohe Stabilität. Niedrige Kosten. |
|
| High-Fidelity Pro-Modell. |
|
| Alias für Abwärtskompatibilität. |
|
| Alias für Abwärtskompatibilität. |
Parameter
Vollständige Parameterreferenz für das generate_image-Tool.
Parameter | Typ | Standard | Beschreibung |
| string | erforderlich | Beschreibung des zu generierenden Bildes |
| string | erforderlich | Ausgabedateiname (Erweiterung wird automatisch hinzugefügt, falls fehlend) |
| enum |
|
|
| enum |
| Modell: |
| enum | auto |
|
| boolean |
| Überspringt die Größenänderung nach der Generierung und gibt native Gemini-Dimensionen zurück |
| enum | optional | Erforderlich, wenn |
| integer | erforderlich außer | Endgültige Ausgabebreite in Pixeln |
| integer | erforderlich außer | Endgültige Ausgabehöhe in Pixeln |
| enum |
|
|
| string | erforderlich für | Absoluter Ausgabeverzeichnispfad |
| boolean |
| Transparenter Hintergrund (nur PNG/WebP) |
| string oder null |
| Farbschlüssel-Überschreibung für Transparenzextraktion |
| integer |
| Toleranz für die Übereinstimmung der Transparenzfarbe |
| enum |
|
|
| enum |
|
|
| enum |
|
|
| enum |
|
|
| boolean |
| Gibt Modell-Reasoning-Felder zurück, wenn Metadaten aktiviert sind |
| boolean |
| Beinhaltet Grounding- und Reasoning-Metadaten in der JSON-Ausgabe |
| array |
| Bis zu 14 lokale Referenzdateien (Flash3.1/Pro3), 3 für Flash2.5 |
| boolean |
| Speichert zwischenzeitliche Debug-Artefakte |
Warum alphabanana?
Keine Wasserzeichen: API-native saubere Bilder.
Thinking/Grounding-Unterstützung: Höhere Prompt-Einhaltung und suchbasierte Genauigkeit.
Produktionsbereit: Unterstützt transparentes WebP und exakte Seitenverhältnisse für Web- und Spiele-Assets.
Funktionen
Ultraschnelle Bildgenerierung (Gemini 3.1 Flash, 0.5K/1K/2K/4K)
Erweitertes Multi-Bild-Reasoning (bis zu 14 Referenzbilder)
Thinking/Grounding-Unterstützung (nur Flash3.1)
Transparente PNG/WebP-Ausgabe (Farbschlüssel-Nachbearbeitung, Despill)
Mehrere Ausgabeformate: Datei, Base64 oder beides
Flexible Größenänderungsmodi: crop, stretch, letterbox, contain
Mehrere Modellstufen: Flash3.1, Flash2.5, Pro3, Legacy-Aliase
Beispielausgaben
Diese Beispielausgaben wurden mit mcp-alphabanana generiert und in images/examples gespeichert.
Pixel-Art-Asset | Referenzbild-Spielszene | Fotorealistische Generierung |
|
|
|
Konfiguration
Konfigurieren Sie den GEMINI_API_KEY in Ihrer MCP-Konfiguration (zum Beispiel mcp.json).
Beispiele:
Referenzieren Sie eine OS-Umgebungsvariable aus
mcp.json:
{
"env": {
"GEMINI_API_KEY": "${env:GEMINI_API_KEY}"
}
}Geben Sie den Schlüssel direkt in
mcp.jsonan:
{
"env": {
"GEMINI_API_KEY": "your_api_key_here"
}
}VS Code Integration
Fügen Sie dies zu Ihren VS Code-Einstellungen (.vscode/settings.json oder Benutzereinstellungen) hinzu und konfigurieren Sie den Server env in mcp.json oder über die VS Code MCP-Einstellungen.
{
"mcp": {
"servers": {
"mcp-alphabanana": {
"command": "npx",
"args": ["-y", "@tasopen/mcp-alphabanana"],
"env": {
"GEMINI_API_KEY": "${env:GEMINI_API_KEY}"
}
}
}
}
}Optional: Legen Sie ein benutzerdefiniertes Fallback-Verzeichnis für Schreibfehler fest, indem Sie MCP_FALLBACK_OUTPUT zum env-Objekt hinzufügen.
Nutzungsbeispiele
Grundlegende Generierung
{
"prompt": "A pixel art treasure chest, golden trim, wooden texture",
"model": "Flash3.1",
"outputFileName": "chest",
"outputType": "base64",
"outputWidth": 64,
"outputHeight": 64,
"transparent": true
}Native Größe ohne Größenänderung
{
"prompt": "A clean app icon with a banana mascot, flat graphic design",
"model": "Flash3.1",
"outputFileName": "banana-icon-native",
"outputType": "base64",
"noresize": true,
"aspectRatio": "1:1",
"output_resolution": "0.5K",
"output_format": "png"
}Dieser Modus gibt die native Gemini-Pixelgröße für das angeforderte Verhältnis und die Auflösung zurück. Zum Beispiel gibt 1:1 + 0.5K 512x512 ohne Größenänderungsschritt zurück.
Erweitert (Vertikales Poster und Thinking)
{
"prompt": "A vertical, photorealistic travel poster advertising Magical Wings Day Tours. A joyful young couple flies high above a breathtaking European countryside at golden hour, holding hands as they soar through a partly cloudy sky. Below them are vineyards, villages, forests, a winding river, and a hilltop medieval castle. The poster uses large, elegant typography with the headline FLY THE COUNTRYSIDE at the top and Magical Wings Day Tours branding near the bottom.",
"model": "Flash3.1",
"output_resolution": "1K",
"outputFileName": "photoreal-travel-poster",
"outputType": "file",
"outputPath": "/path/to/output",
"outputWidth": 848,
"outputHeight": 1264,
"output_format": "jpg",
"thinking_mode": "high",
"include_metadata": true
}Grounding-Beispiel (Suchbasiert)
{
"prompt": "A modern travel poster featuring today's weather and skyline highlights in Kuala Lumpur",
"model": "Flash3.1",
"outputFileName": "kl_travel_poster",
"outputType": "base64",
"outputWidth": 1024,
"outputHeight": 1024,
"grounding_type": "text",
"thinking_mode": "high",
"include_metadata": true,
"include_thoughts": true
}Dieses Beispiel aktiviert das Google-Suche-Grounding und gibt Grounding- und Reasoning-Metadaten in JSON zurück.
Mit Referenzbildern
{
"prompt": "Use the reference image to create a game screen showing an opened treasure chest filled with coins and treasure, 8-bit dungeon crawler style, after-battle reward scene, dungeon corridor background, four-party status UI at the bottom",
"model": "Flash3.1",
"output_resolution": "0.5K",
"outputFileName": "reference-image-dungeon-loot",
"outputType": "file",
"outputPath": "/path/to/output",
"outputWidth": 600,
"outputHeight": 448,
"output_format": "webp",
"transparent": false,
"referenceImages": [
{
"description": "Treasure chest style reference",
"filePath": "/path/to/references/pixel-art-treasure-chest.png"
}
]
}Transparenz & Ausgabeformate
PNG: Volle Alpha, Farbschlüssel + Despill
WebP: Volle Alpha, bessere Komprimierung (Flash3.1+)
JPEG: Keine Transparenz (fällt auf soliden Hintergrund zurück)
Entwicklung
# Development mode with MCP CLI
npm run dev
# MCP Inspector (Web UI)
npm run inspect
# Build for production
npm run buildLizenz
MIT
Available Tools
1 toolgenerate_imageADestructive
Generate image assets using Gemini AI with optional transparency and reference images.
[Claude Desktop Guidance]
Prefer outputType='file' for medium or large images. base64 and combine responses can exceed Claude Desktop's context limit.
On Claude Desktop for Windows, use the FileSystem extension to choose reference-image paths and a writable absolute outputPath before calling this tool.
Use base64 only for small previews or when the client explicitly needs inline image data.
[Model Guidance]
NanoBanana2.1 (default): Google Nano Banana 2.1 (gemini-nano-banana-2.1). Best overall quality, prompt adherence, character consistency, text rendering and reference-image handling. Supports 1K/2K/4K and up to 14 reference images, plus thinking and grounding.
Flash3.1 (deprecated): gemini-3.1-flash-image. Google shuts it down on October 29, 2026; requests are rejected from that date. Please migrate to NanoBanana2.1.
Lite3.1 (Nano Banana 2 Lite): Ultra-fast, cost-effective, 1K-only, no search grounding. Ideal for quick drafting and low-latency iteration.
Pro3: Higher fidelity, but more costly and slower.
Flash2.5: Legacy, maintained for compatibility. 1K only. Aliases: 'flash' resolves to NanoBanana2.1 (the default), 'pro' to Pro3.
[Thinking] thinking_mode accepts minimal, medium, high (default: medium, matching NanoBanana2.1's model default). NanoBanana2.1 supports all three levels; Flash3.1 and Pro3 support minimal/high and fall back to their own default when given an unsupported level.
[Resolutions] output_resolution supports 1K, 2K and 4K (0.5K was removed in v1.6.0). It is auto-selected from the requested pixel size in normal mode.
[Aspect Ratios] Gemini supports the following aspect ratios (model-dependent):
Common to all models: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9
Flash3.1 only: 1:4, 4:1, 1:8, 8:1
Normal mode: provide outputWidth/outputHeight and the server will choose the closest Gemini aspect ratio and source resolution, then resize to the requested pixel size. No-resize mode: set noresize=true and provide aspectRatio plus output_resolution. The server will return Gemini's native pixel dimensions for that combination without post-generation resizing.
If you intentionally want to control resizing/cropping in normal mode, use the 'resizeMode' parameter: 'crop' (default, center crop), 'letterbox' (fit with padding), 'contain' (trim transparent margins then fit), or 'stretch' (distort to fit).
[IMPORTANT] Always preserve the user's prompt as-is, including language and nuance. Do not translate or summarize.
| Name | Required | Description | Default |
|---|---|---|---|
| debug | No | Debug mode: output intermediate processing images and prompt | |
| model | No | Model tier to use for generation (default: NanoBanana2.1; see tool description for details; "flash" resolves to NanoBanana2.1, "pro" to Pro3; "Lite3.1" is the low-latency Nano Banana 2 Lite model, 1K-only, no grounding; "Flash3.1" is deprecated — Google shuts it down on October 29, 2026) | NanoBanana2.1 |
| prompt | Yes | User-provided image prompt. Preserve the original wording and detail; do not summarize or translate. Only append transparency-related hints if needed. | |
| noresize | No | Skip post-generation resizing and return Gemini native dimensions directly. When true, provide aspectRatio and output_resolution instead of outputWidth/outputHeight. | |
| fringeMode | No | Fringe reduction mode: auto (size-based), crisp (binary alpha), hd (force-clear 1px boundary for large images). | auto |
| outputPath | No | Output directory path (MUST be an absolute path when outputType is file or combine). In Claude Desktop on Windows, use the FileSystem extension to choose or prepare a writable absolute path such as C:\temp. | |
| outputType | No | Output format: file=file only, base64=base64 only, combine=both. In Claude Desktop, prefer file for medium or large images to avoid context-size limits; use base64 only for small previews. | combine |
| resizeMode | No | Resize mode: crop=center crop, stretch=distort, letterbox=fit with padding, contain=trim transparent margins then fit | crop |
| aspectRatio | No | Gemini aspect ratio to use directly when noresize=true. Ignored in normal resize mode. | |
| outputWidth | No | Output image width in pixels. Required unless noresize=true. In normal mode, the image will be generated using the closest supported Gemini aspect ratio and resolution, then resized to this width. | |
| transparent | No | Request transparent background (PNG or WebP only). Background color is selected by histogram analysis. | |
| outputHeight | No | Output image height in pixels. Required unless noresize=true. In normal mode, the image will be generated using the closest supported Gemini aspect ratio and resolution, then resized to this height. | |
| output_format | No | Output format | png |
| thinking_mode | No | Thinking level: minimal, medium, high (default: medium, matching NanoBanana2.1's model default; requires a thinking-capable model — NanoBanana2.1 accepts all three levels, Flash3.1/Pro3 accept minimal/high and fall back to their own default otherwise) | medium |
| colorTolerance | No | Tolerance for color matching (0-255). Higher values are more permissive for transparent color selection and keying. | |
| grounding_type | No | Grounding tool usage (requires a grounding-capable model; see [Model Guidance] in the tool description) | none |
| outputFileName | Yes | Output filename (extension auto-added if missing) | |
| referenceImages | No | Reference images for style guidance (max 14 files; per-model limits are enforced automatically) | |
| include_metadata | No | Include grounding and reasoning metadata in JSON output (optional, may increase payload size). | |
| include_thoughts | No | Optional (default: false). Request thought fields from Gemini (thinking-capable models only). Thought content is returned in MCP response only when include_metadata=true. | |
| transparentColor | No | Color to make transparent. Hex (e.g. #FF00FF). null defaults to #FF00FF when transparent=true. | |
| output_resolution | No | Gemini generation source resolution: 1K, 2K, and 4K (optional in normal mode, required when noresize=true; 0.5K was removed in v1.6.0). In normal mode, the final image is resized to the requested pixel size after generation. |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations declare destructiveHint=true and openWorldHint=true, and the description adds substantial behavioral context beyond them: model deprecation dates, fallback behavior for unsupported thinking levels, histogram-based background selection, and the auto-resize pipeline. It does not explain the destructive aspect — notably whether an existing file at outputPath is overwritten — which is the main remaining gap.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Front-loaded purpose followed by clearly bracketed sections, so the agent can scan to the relevant concern. It is long and the [Thinking], [Resolutions], and [Aspect Ratios] sections restate schema parameter descriptions, so not every sentence is strictly additive.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a 22-parameter tool with no output schema, the description covers client constraints, model tiers, mode interplay, and prompt-preservation rules. It leaves unstated what happens on unwritable outputPath or on overwrite of an existing file, which matters given destructiveHint=true.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline is 3, but the description adds model-dependent semantics not in the schema: which aspect ratios are Flash3.1-only, that aliases resolve ('flash' to NanoBanana2.1, 'pro' to Pro3), and how normal vs noresize mode changes which of outputWidth/outputHeight, aspectRatio, and output_resolution matter.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The opening sentence states a specific verb and resource ('Generate image assets using Gemini AI') plus the two distinguishing capabilities (optional transparency, reference images). With no sibling tools to disambiguate against, this fully identifies the operation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives explicit when-to-use rules: prefer outputType='file' for medium/large images, base64 only for small previews, use FileSystem extension on Windows for reference paths, and normal mode vs noresize mode selection. Model selection is also routed with named tradeoffs (Lite3.1 for drafting, Pro3 for fidelity).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
1 tool update
v1.6.0- Changed
generate_image13 fields changed- changed
Input schema / properties / grounding_type / descriptionPrevious value: -"Grounding tool usage (3.1 only)"New value: +"Grounding tool usage (requires a grounding-capable model; see [Model Guidance] in the tool description)" - changed
Input schema / properties / include_thoughts / descriptionPrevious value: -"Optional (default: false). Request thought fields from Gemini (3.1 only). Thought content is returned in MCP response only when include_metadata=true."New value: +"Optional (default: false). Request thought fields from Gemini (thinking-capable models only). Thought content is returned in MCP response only when include_metadata=true." - changed
Input schema / properties / model / defaultPrevious value: -"Flash3.1"New value: +"NanoBanana2.1" - changed
Input schema / properties / model / descriptionPrevious value: -"Model tier to use for generation (see tool description for details; \"flash\" and \"pro\" are aliases for Flash2.5 and Pro3; \"Lite3.1\" is the low-latency Nano Banana 2 Lite model, 1K-only, no grounding)"New value: +"Model tier to use for generation (default: NanoBanana2.1; see tool description for details; \"flash\" resolves to NanoBanana2.1, \"pro\" to Pro3; \"Lite3.1\" is the low-latency Nano Banana 2 Lite model, 1K-only, no grounding; \"Flash3.1\" is deprecated — Google shuts it down on October 29, 2026)" - changed
Input schema / properties / model / enumPrevious value: -[ - "Flash3.1", - "Lite3.1", - "Flash2.5", - "Pro3", - "flash", - "pro" -]New value: +[ + "NanoBanana2.1", + "Flash3.1", + "Lite3.1", + "Flash2.5", + "Pro3", + "flash", + "pro" +] - changed
Input schema / properties / output_resolution / descriptionPrevious value: -"Gemini generation source resolution (optional in normal mode, required when noresize=true). In normal mode, the final image is resized to the requested pixel size after generation."New value: +"Gemini generation source resolution: 1K, 2K, and 4K (optional in normal mode, required when noresize=true; 0.5K was removed in v1.6.0). In normal mode, the final image is resized to the requested pixel size after generation." - removed
Input schema / properties / output_resolution / enumRemoved value: -[ - "0.5K", - "1K", - "2K", - "4K" -] - changed
Input schema / properties / referenceImages / descriptionPrevious value: -"Reference images for style guidance (Flash2.5: max 3, others: max 14)"New value: +"Reference images for style guidance (max 14 files; per-model limits are enforced automatically)" - changed
Input schema / properties / thinking_mode / defaultPrevious value: -"minimal"New value: +"medium" - changed
Input schema / properties / thinking_mode / descriptionPrevious value: -"Thinking mode (3.1 only)"New value: +"Thinking level: minimal, medium, high (default: medium, matching NanoBanana2.1's model default; requires a thinking-capable model — NanoBanana2.1 accepts all three levels, Flash3.1/Pro3 accept minimal/high and fall back to their own default otherwise)" - changed
Input schema / properties / thinking_mode / enumPrevious value: -[ - "minimal", - "high" -]New value: +[ + "minimal", + "medium", + "high" +] - removed
Input schema / properties / transparentColor / anyOfRemoved value: -[ - { - "type": "string" - }, - { - "type": "null" - } -] - added
Input schema / properties / transparentColor / typeAdded value: +[ + "string", + "null" +]
1 tool update
v1.5.0- Changed
generate_image14 fields changed- added
Input schema / properties / aspectRatioAdded value: +{ + "description": "Gemini aspect ratio to use directly when noresize=true. Ignored in normal resize mode.", + "enum": [ + "1:1", + "2:3", + "3:2", + "3:4", + "4:3", + "4:5", + "5:4", + "9:16", + "16:9", + "21:9", + "1:4", + "4:1", + "1:8", + "8:1" + ], + "type": "string" +} - changed
Input schema / properties / model / descriptionPrevious value: -"Model tier to use for generation (see tool description for details; \"flash\" and \"pro\" are aliases for Flash2.5 and Pro3)"New value: +"Model tier to use for generation (see tool description for details; \"flash\" and \"pro\" are aliases for Flash2.5 and Pro3; \"Lite3.1\" is the low-latency Nano Banana 2 Lite model, 1K-only, no grounding)" - changed
Input schema / properties / model / enumPrevious value: -[ - "Flash3.1", - "Flash2.5", - "Pro3", - "flash", - "pro" -]New value: +[ + "Flash3.1", + "Lite3.1", + "Flash2.5", + "Pro3", + "flash", + "pro" +] - added
Input schema / properties / noresizeAdded value: +{ + "default": false, + "description": "Skip post-generation resizing and return Gemini native dimensions directly. When true, provide aspectRatio and output_resolution instead of outputWidth/outputHeight.", + "type": "boolean" +} - changed
Input schema / properties / outputHeight / descriptionPrevious value: -"Output image height in pixels. The image will be generated using the closest supported Gemini aspect ratio and resolution, then resized to this height. To avoid cropping or padding, set width and height to match a supported aspect ratio (see tool description)."New value: +"Output image height in pixels. Required unless noresize=true. In normal mode, the image will be generated using the closest supported Gemini aspect ratio and resolution, then resized to this height." - changed
Input schema / properties / outputPath / descriptionPrevious value: -"Output directory path (MUST be an absolute path when outputType is file or combine)"New value: +"Output directory path (MUST be an absolute path when outputType is file or combine). In Claude Desktop on Windows, use the FileSystem extension to choose or prepare a writable absolute path such as C:\\temp." - changed
Input schema / properties / outputType / descriptionPrevious value: -"Output format: file=file only, base64=base64 only, combine=both"New value: +"Output format: file=file only, base64=base64 only, combine=both. In Claude Desktop, prefer file for medium or large images to avoid context-size limits; use base64 only for small previews." - changed
Input schema / properties / outputWidth / descriptionPrevious value: -"Output image width in pixels. The image will be generated using the closest supported Gemini aspect ratio and resolution, then resized to this width. To avoid cropping or padding, set width and height to match a supported aspect ratio (see tool description)."New value: +"Output image width in pixels. Required unless noresize=true. In normal mode, the image will be generated using the closest supported Gemini aspect ratio and resolution, then resized to this width." - changed
Input schema / properties / output_resolution / descriptionPrevious value: -"Gemini generation source resolution (optional; normally auto-calculated from pixel size. Set only to override. Final image is resized to requested pixel size.)"New value: +"Gemini generation source resolution (optional in normal mode, required when noresize=true). In normal mode, the final image is resized to the requested pixel size after generation." - removed
Input schema / properties / referenceImages / items / additionalPropertiesRemoved value: -false - changed
Input schema / properties / referenceImages / items / properties / filePath / descriptionPrevious value: -"Absolute path to reference image file (.png, .jpg, .jpeg, .webp)"New value: +"Absolute path to reference image file (.png, .jpg, .jpeg, .webp). In Claude Desktop on Windows, use the FileSystem extension to locate the file and pass its Windows absolute path." - added
Input schema / properties / transparentColor / anyOfAdded value: +[ + { + "type": "string" + }, + { + "type": "null" + } +] - removed
Input schema / properties / transparentColor / typeRemoved value: -[ - "string", - "null" -] - changed
Input schema / requiredPrevious value: -[ - "prompt", - "outputFileName", - "outputWidth", - "outputHeight" -]New value: +[ + "prompt", + "outputFileName" +]
1 tool update
v1.3.6- First observed
generate_image
TDQS
Scored across 1 tool
With a single tool there is no possibility of misselection; generate_image has one unambiguous purpose. The extensive guidance distinguishes modes and parameters rather than competing tools, so no disambiguation burden is placed on the agent.
generate_image follows a clear verb_noun snake_case pattern. There are no other names to conflict with, so consistency is trivially satisfied and readable.
A one-tool surface is thin for what is effectively a full image-generation service with multiple models, resolutions, reference handling, and resize modes. Much of that surface is expressed through parameters and documentation rather than discoverable tools, making the count borderline.
Core generation is well covered: model choice, resolution, aspect ratio, transparency, reference images, and resize behavior are all parameterized. Minor gaps exist around ancillary operations such as editing/upscaling existing assets or programmatically listing available models, which are only described in prose guidance.
Maintenance
Related MCP Connectors
MCP server for Qwen Image 3 AI image generation
Focused MCP server for OpenAI image/audio generation (v2.0.0). Wraps endpoints via HAPI CLI.
MCP server for Google Veo AI video generation
MCP server for Hailuo (MiniMax) AI video generation
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceMCP server for generating, editing, and processing images via multiple providers including Kilo, OpenRouter, OpenAI, and Gemini, with local tools for background removal, resizing, and cropping.57 npm3MIT
- AlicenseNot gradedqualityDmaintenanceAn MCP server that provides image generation using Google's Nano Banana Gemini models, with additional tools for background removal, upscaling, and format conversion via deterministic post-processing.1MIT
- FlicenseAqualityDmaintenanceWraps Google Gemini's image generation API as an MCP server, enabling text-to-image, image editing, and grounded search workflows from any MCP client.2-
- AlicenseNot gradedqualityDmaintenanceMCP server for Google's Nano Banana image generation models (Gemini). Generate and edit images via natural language.12 npm1MIT

