mcp-hydrocoder-image 1.1.1 → 1.2.1

package/README.md

@@ -1,6 +1,6 @@
 # MCP HydroCoder Image 🍌
 
-> AI image generation and editing MCP server for Cursor, Claude Code, Codex, and any MCP-compatible tool — powered by Nano Banana 2 and Nano Banana Pro (Google Gemini).
+> AI image generation and editing MCP server for Cursor, Claude Code, Codex, and any MCP-compatible tool — powered by Volcengine Seedream and Google Gemini.
 
 [![npm version](https://badge.fury.io/js/mcp-hydrocoder-image.svg)](https://www.npmjs.com/package/mcp-hydrocoder-image)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
@@ -36,6 +36,10 @@ The prompt optimizer uses a **Subject–Context–Style** framework (powered by
 
 ## Features
 
+- **Multi-Provider Support**: Use either Volcengine Seedream or Google Gemini behind the same MCP server, with provider selection via config or per-request override.
+- **Provider-Aware Routing**: If the caller explicitly asks for Volcengine / Seedream / 豆包 / 火山引擎 or Gemini / Nano Banana, the server metadata now nudges agents to pass the correct `provider`, and the server can infer the provider from explicit prompt instructions when the parameter is omitted.
+- **Volcengine-First Production Path**: README and install examples now default to Volcengine because it is the more stable day-to-day path for text-to-image and grouped output in this project.
+- **Dedicated Multi-Image Tooling**: Use `generate_image` for single-image work and `generate_multi_image` for grouped multi-image output with server-side fan-out and prompt normalization.
 - **Built-in Prompt Optimization**: Your simple prompt is automatically enriched with photographic and artistic details — lighting, composition, atmosphere — using Gemini 2.5 Flash. No prompt engineering skills required.
 - **Three Quality Tiers**: Choose between fast iteration, balanced quality, or maximum fidelity with Nano Banana 2 (Gemini 3.1 Flash Image) and Nano Banana Pro (Gemini 3 Pro Image). [See Quality Presets](#quality-presets).
 - **Image Editing**: Transform existing images with natural language instructions (image-to-image) while preserving original style and visual consistency.
@@ -81,24 +85,30 @@ npx mcp-hydrocoder-image skills install --path ~/.claude/skills
 | | MCP Server | Agent Skill |
 |---|---|---|
 | **Use when** | Your AI tool does not have built-in image generation | Your AI tool already generates images natively |
-| **Requires** | Gemini API key | Nothing |
-| **What it does** | Generates images via Gemini API with automatic prompt optimization | Teaches the AI to write better prompts |
+| **Requires** | Volcengine or Gemini API key | Nothing |
+| **What it does** | Generates images via Volcengine or Gemini API with automatic prompt optimization | Teaches the AI to write better prompts |
 | **Works with** | MCP-compatible tools (Cursor, Claude Code, Codex, etc.) | Any tool supporting the [Agent Skills](https://agentskills.io) open standard |
 
 ---
 
 ## Prerequisites
 
-- **Node.js** 20 or higher
-- **Gemini API Key** - Get yours at [Google AI Studio](https://aistudio.google.com/apikey)
-- An MCP-compatible AI tool: **Cursor**, **Claude Code**, **Codex**, or others
-- Basic terminal/command line knowledge
+- **Node.js** 20 or higher
+- **API Key**
+  - Volcengine: create an Ark API key in the Volcengine console
+  - Gemini: get yours at [Google AI Studio](https://aistudio.google.com/apikey)
+- An MCP-compatible AI tool: **Cursor**, **Claude Code**, **Codex**, or others
+- Basic terminal/command line knowledge
 
 ## Quick Start
 
-### 1. Get Your Gemini API Key
+### 1. Get Your API Key
 
-Get your API key from [Google AI Studio](https://aistudio.google.com/apikey)
+Choose a provider and create an API key:
+- **Volcengine**: Volcengine Ark console (`ARK_API_KEY`-compatible key)
+- **Gemini**: [Google AI Studio](https://aistudio.google.com/apikey)
+
+> **Recommended default**: Start with Volcengine. In this project it is the preferred production path, especially for single-image generation, grouped multi-image workflows, and predictable output sizing.
 
 ### 2. MCP Configuration
 
@@ -111,11 +121,14 @@ Add to `~/.codex/config.toml`:
 command = "npx"
 args = ["-y", "mcp-hydrocoder-image"]
 
-[mcp_servers.mcp-hydrocoder-image.env]
-GEMINI_API_KEY = "your_gemini_api_key_here"
-IMAGE_OUTPUT_DIR = "/absolute/path/to/images"
-API_TIMEOUT = "120000"  # Optional: timeout in milliseconds (default: 120s)
-```
+[mcp_servers.mcp-hydrocoder-image.env]
+IMAGE_PROVIDER = "volcengine"
+VOLCENGINE_API_KEY = "your_volcengine_api_key_here"
+VOLCENGINE_MODEL = "doubao-seedream-4-5-251128"
+VOLCENGINE_API_BASE_URL = "https://ark.cn-beijing.volces.com/api/v3"
+IMAGE_OUTPUT_DIR = "/absolute/path/to/images"
+API_TIMEOUT = "120000"  # Optional: timeout in milliseconds (default: 120s)
+```
 
 #### For Cursor
 
@@ -128,13 +141,16 @@ Add to your Cursor settings:
 {
   "mcpServers": {
     "mcp-hydrocoder-image": {
-      "command": "npx",
-      "args": ["-y", "mcp-hydrocoder-image"],
-      "env": {
-        "GEMINI_API_KEY": "your_gemini_api_key_here",
-        "IMAGE_OUTPUT_DIR": "/absolute/path/to/images",
-        "API_TIMEOUT": "120000"
-      }
+      "command": "npx",
+      "args": ["-y", "mcp-hydrocoder-image"],
+      "env": {
+        "IMAGE_PROVIDER": "volcengine",
+        "VOLCENGINE_API_KEY": "your_volcengine_api_key_here",
+        "VOLCENGINE_MODEL": "doubao-seedream-4-5-251128",
+        "VOLCENGINE_API_BASE_URL": "https://ark.cn-beijing.volces.com/api/v3",
+        "IMAGE_OUTPUT_DIR": "/absolute/path/to/images",
+        "API_TIMEOUT": "120000"
+      }
     }
   }
 }
@@ -145,13 +161,16 @@ Add to your Cursor settings:
 {
   "mcpServers": {
     "mcp-hydrocoder-image": {
-      "command": "npx",
-      "args": ["-y", "mcp-hydrocoder-image"],
-      "env": {
-        "GEMINI_API_KEY": "your_gemini_api_key_here",
-        "IMAGE_OUTPUT_DIR": "C:\\absolute\\path\\to\\images",
-        "API_TIMEOUT": "120000"
-      }
+      "command": "npx",
+      "args": ["-y", "mcp-hydrocoder-image"],
+      "env": {
+        "IMAGE_PROVIDER": "volcengine",
+        "VOLCENGINE_API_KEY": "your_volcengine_api_key_here",
+        "VOLCENGINE_MODEL": "doubao-seedream-4-5-251128",
+        "VOLCENGINE_API_BASE_URL": "https://ark.cn-beijing.volces.com/api/v3",
+        "IMAGE_OUTPUT_DIR": "C:\\absolute\\path\\to\\images",
+        "API_TIMEOUT": "120000"
+      }
     }
   }
 }
@@ -163,21 +182,27 @@ Run in your project directory to enable for that project:
 
 ```bash
 cd /path/to/your/project
-claude mcp add mcp-hydrocoder-image \
-  --env GEMINI_API_KEY=your-api-key \
-  --env IMAGE_OUTPUT_DIR=/absolute/path/to/images \
-  --env API_TIMEOUT=120000 \
-  -- npx -y mcp-hydrocoder-image
+claude mcp add mcp-hydrocoder-image \
+  --env IMAGE_PROVIDER=volcengine \
+  --env VOLCENGINE_API_KEY=your-volcengine-api-key \
+  --env VOLCENGINE_MODEL=doubao-seedream-4-5-251128 \
+  --env VOLCENGINE_API_BASE_URL=https://ark.cn-beijing.volces.com/api/v3 \
+  --env IMAGE_OUTPUT_DIR=/absolute/path/to/images \
+  --env API_TIMEOUT=120000 \
+  -- npx -y mcp-hydrocoder-image
 ```
 
 Or add globally for all projects:
 
 ```bash
-claude mcp add mcp-hydrocoder-image --scope user \
-  --env GEMINI_API_KEY=your-api-key \
-  --env IMAGE_OUTPUT_DIR=/absolute/path/to/images \
-  --env API_TIMEOUT=120000 \
-  -- npx -y mcp-hydrocoder-image
+claude mcp add mcp-hydrocoder-image --scope user \
+  --env IMAGE_PROVIDER=volcengine \
+  --env VOLCENGINE_API_KEY=your-volcengine-api-key \
+  --env VOLCENGINE_MODEL=doubao-seedream-4-5-251128 \
+  --env VOLCENGINE_API_BASE_URL=https://ark.cn-beijing.volces.com/api/v3 \
+  --env IMAGE_OUTPUT_DIR=/absolute/path/to/images \
+  --env API_TIMEOUT=120000 \
+  -- npx -y mcp-hydrocoder-image
 ```
 
 Or add via JSON config (`~/.claude/settings.json` for global, `.mcp.json` for project):
@@ -187,13 +212,16 @@ Or add via JSON config (`~/.claude/settings.json` for global, `.mcp.json` for pr
 {
   "mcpServers": {
     "mcp-hydrocoder-image": {
-      "command": "npx",
-      "args": ["-y", "mcp-hydrocoder-image"],
-      "env": {
-        "GEMINI_API_KEY": "your_gemini_api_key_here",
-        "IMAGE_OUTPUT_DIR": "/absolute/path/to/images",
-        "API_TIMEOUT": "120000"
-      }
+      "command": "npx",
+      "args": ["-y", "mcp-hydrocoder-image"],
+      "env": {
+        "IMAGE_PROVIDER": "volcengine",
+        "VOLCENGINE_API_KEY": "your_volcengine_api_key_here",
+        "VOLCENGINE_MODEL": "doubao-seedream-4-5-251128",
+        "VOLCENGINE_API_BASE_URL": "https://ark.cn-beijing.volces.com/api/v3",
+        "IMAGE_OUTPUT_DIR": "/absolute/path/to/images",
+        "API_TIMEOUT": "120000"
+      }
     }
   }
 }
@@ -204,26 +232,88 @@ Or add via JSON config (`~/.claude/settings.json` for global, `.mcp.json` for pr
 {
   "mcpServers": {
     "mcp-hydrocoder-image": {
-      "command": "npx",
-      "args": ["-y", "mcp-hydrocoder-image"],
-      "env": {
-        "GEMINI_API_KEY": "your_gemini_api_key_here",
-        "IMAGE_OUTPUT_DIR": "C:\\absolute\\path\\to\\images",
-        "API_TIMEOUT": "120000"
-      }
+      "command": "npx",
+      "args": ["-y", "mcp-hydrocoder-image"],
+      "env": {
+        "IMAGE_PROVIDER": "volcengine",
+        "VOLCENGINE_API_KEY": "your_volcengine_api_key_here",
+        "VOLCENGINE_MODEL": "doubao-seedream-4-5-251128",
+        "VOLCENGINE_API_BASE_URL": "https://ark.cn-beijing.volces.com/api/v3",
+        "IMAGE_OUTPUT_DIR": "C:\\absolute\\path\\to\\images",
+        "API_TIMEOUT": "120000"
+      }
     }
   }
 }
 ```
 
-⚠️ **Security Note**: Never commit your API key to version control. Keep it secure and use environment-specific configuration.
-
-📁 **Path Requirements**:
-- `IMAGE_OUTPUT_DIR` must be an absolute path (e.g., `/Users/username/images`, not `./images`)
-- Defaults to `./output` in the current working directory if not specified
-- Directory will be created automatically if it doesn't exist
-
-#### Custom API Base URL (Third-party Proxy)
+⚠️ **Security Note**: Never commit your API key to version control. Keep it secure and use environment-specific configuration.
+
+📁 **Path Requirements**:
+- `IMAGE_OUTPUT_DIR` must be an absolute path (e.g., `/Users/username/images`, not `./images`)
+- Defaults to `./output` in the current working directory if not specified
+- Directory will be created automatically if it doesn't exist
+
+#### Volcengine Local MCP Example
+
+For local MCP testing, build the project first and point your MCP client at the local `dist/index.js` entry instead of `npx`.
+
+```bash
+npm install
+npm run build
+```
+
+**Codex (`~/.codex/config.toml` on Windows):**
+```toml
+[mcp_servers.mcp-hydrocoder-image-local]
+command = "node"
+args = ["C:\\workspace\\develop\\ccExtensions\\mcpBanana\\dist\\index.js"]
+
+[mcp_servers.mcp-hydrocoder-image-local.env]
+IMAGE_PROVIDER = "volcengine"
+VOLCENGINE_API_KEY = "your_volcengine_api_key_here"
+VOLCENGINE_MODEL = "doubao-seedream-4-5-251128"
+VOLCENGINE_API_BASE_URL = "https://ark.cn-beijing.volces.com/api/v3"
+IMAGE_OUTPUT_DIR = "C:\\workspace\\develop\\ccExtensions\\mcpBanana\\output"
+API_TIMEOUT = "120000"
+```
+
+**Cursor / Claude Code JSON config (Windows local repo example):**
+```json
+{
+  "mcpServers": {
+    "mcp-hydrocoder-image-local": {
+      "command": "node",
+      "args": [
+        "C:\\workspace\\develop\\ccExtensions\\mcpBanana\\dist\\index.js"
+      ],
+      "env": {
+        "IMAGE_PROVIDER": "volcengine",
+        "VOLCENGINE_API_KEY": "your_volcengine_api_key_here",
+        "VOLCENGINE_MODEL": "doubao-seedream-4-5-251128",
+        "VOLCENGINE_API_BASE_URL": "https://ark.cn-beijing.volces.com/api/v3",
+        "IMAGE_OUTPUT_DIR": "C:\\workspace\\develop\\ccExtensions\\mcpBanana\\output",
+        "API_TIMEOUT": "120000"
+      }
+    }
+  }
+}
+```
+
+If you move the repo, update both the `args` path and `IMAGE_OUTPUT_DIR` to the new absolute path.
+
+#### Gemini Example
+
+If you prefer Gemini, switch `IMAGE_PROVIDER` to `gemini` and replace the Volcengine variables with `GEMINI_API_KEY`. Gemini remains supported, but the README now treats it as the optional path rather than the default install path.
+
+For standard direct access to the official Gemini endpoint, you only need:
+- `IMAGE_PROVIDER=gemini`
+- `GEMINI_API_KEY=your_gemini_api_key_here`
+- `IMAGE_OUTPUT_DIR=/absolute/path/to/images`
+
+`GEMINI_API_BASE_URL` is optional. Add it only when you use a third-party proxy, relay, or custom compatible gateway.
+
+#### Gemini Custom API Base URL (Third-party Proxy)
 
 To use a third-party API endpoint or proxy, add the `GEMINI_API_BASE_URL` environment variable:
 
@@ -263,7 +353,19 @@ IMAGE_OUTPUT_DIR = "/absolute/path/to/images"
 
 > **Note**: The base URL should be the root domain (e.g., `https://llm.myseek.fun`), without the `/v1` suffix — the SDK will append the API version automatically.
 
-## Quality Presets
+### Volcengine Notes
+
+Volcengine is the recommended default provider in this project.
+
+Current implementation status and behavior:
+- Stable path: text-to-image
+- Reference-image workflows are wired through the OpenAI-compatible image API using the `image` field
+- Base64 image inputs are normalized to the official Volcengine format: `data:image/<format>;base64,<Base64编码>`
+- Grouped output (`outputCount`) is best-effort and depends on provider-side behavior
+- If the user does not specify `aspectRatio` or `imageSize`, Gemini and Volcengine default to `16:9` and `4K`
+- If the user specifies `aspectRatio` and/or `imageSize`, the server automatically normalizes the final `WxH` into Volcengine's legal pixel range
+- When the user provides local image paths, they should be passed through `inputImagePath` / `inputImagePaths` instead of being summarized into the prompt
+
 
 Choose the right balance of speed, quality, and cost:
 
@@ -330,7 +432,7 @@ Your prompt is automatically enhanced with rich details about lighting, material
 (with inputImagePath: "/path/to/image.jpg")
 ```
 
-### Advanced Features
+### Advanced Features
 
 **Character Consistency:**
 ```
@@ -344,41 +446,98 @@ Your prompt is automatically enhanced with rich details about lighting, material
 (with imageSize: "4K")
 ```
 
-**Custom Aspect Ratio:**
-```
-"Generate a cinematic landscape of a desert at golden hour"
-(with aspectRatio: "21:9")
-```
-
-## API Reference
-
-### `generate_image` Tool
+**Custom Aspect Ratio:**
+```
+"Generate a cinematic landscape of a desert at golden hour"
+(with aspectRatio: "21:9")
+```
+
+### Grouped Multi-Image Generation
+
+For a grouped multi-image task, prefer `generate_multi_image` instead of repeating `generate_image`.
+
+**Natural-language grouped request:**
+```text
+"Use generate_multi_image to create 4 unified e-commerce product images of the same minimalist white thermos cup: hero shot, side-detail shot, handheld lifestyle shot, and desk scene. Return 4 separate images in one run."
+```
+
+**Structured grouped request with inferred numbering:**
+```json
+{
+  "prompt": "同一款极简白色保温杯，整体风格统一，高级感、干净、真实、适合品牌官网和详情页使用。",
+  "outputCount": 4,
+  "provider": "volcengine"
+}
+```
+
+**Structured grouped request with explicit per-image prompts:**
+```json
+{
+  "prompt": "同一款极简白色保温杯，整体风格统一，高级感、干净、真实、适合品牌官网和详情页使用。",
+  "provider": "volcengine",
+  "imageRequests": [
+    "电商主图，白底，正面展示产品",
+    "侧面细节图，突出杯盖和材质纹理",
+    "手持使用场景图，突出尺寸感",
+    "办公桌场景图，氛围高级"
+  ]
+}
+```
+
+## API Reference
+
+### `generate_image` Tool
 
 The server uses a two-stage process with separate models for each stage:
 1. **Prompt Optimization** (Gemini 2.5 Flash): Refines your prompt using the Subject–Context–Style framework. Skippable via `SKIP_PROMPT_ENHANCEMENT`.
-2. **Image Generation** (Nano Banana 2 or Pro): Creates the final image. Model varies by quality preset.
+2. **Image Generation** (Nano Banana 2 or Pro): Creates the final image. Model varies by quality preset.
+
+Use `generate_image` as the default tool for single-image generation and image editing. If the user wants multiple images in one grouped request, prefer `generate_multi_image`.
 
 #### Parameters
 
 | Parameter | Type | Required | Description |
 |-----------|------|----------|-------------|
 | `prompt` | string | Yes | Text description or editing instruction |
+| `provider` | string | - | Optional provider override: `gemini` or `volcengine`. Defaults to `IMAGE_PROVIDER`. If omitted, the server can still infer the provider from explicit instructions such as "use Volcengine" / "用火山引擎" / "use Gemini" |
 | `quality` | string | - | Quality preset: `fast` (default), `balanced`, `quality`. Overrides `IMAGE_QUALITY` env var for this request |
-| `inputImagePath` | string | - | Absolute path to input image for image-to-image editing |
-| `inputImage` | string | - | Base64 encoded image data for image-to-image editing. Alternative to `inputImagePath` |
-| `inputImageMimeType` | string | - | MIME type of the input image (`image/jpeg`, `image/png`, `image/webp`, `image/gif`, `image/bmp`). Used with `inputImage` |
-| `inputImages` | array | - | Multiple input images for multi-image composition. Each item: `{ data: string, mimeType: string }`. Cannot be used with `inputImage`/`inputImagePath`/`inputImagePaths` |
-| `inputImagePaths` | array | - | Multiple input image file paths for multi-image composition. Each item is an absolute path string. Cannot be used with other image input params |
+| `outputFormat` | string | - | Output image format if supported by the provider. Some provider endpoints may ignore or reject format overrides |
+| `outputCount` | integer | - | Backward-compatible grouped output count for `generate_image`. For new multi-image requests, prefer `generate_multi_image` |
+| `inputImagePath` | string | - | Absolute path to input image for image-to-image editing. Supported by Gemini and by Volcengine reference-image workflows |
+| `inputImage` | string | - | Base64 encoded image data for image-to-image editing. Gemini accepts raw base64; Volcengine sends it as `data:image/<format>;base64,<data>` and uses `inputImageMimeType` to build the official request format |
+| `inputImageMimeType` | string | - | MIME type of the input image (`image/jpeg`, `image/png`, `image/webp`, `image/gif`, `image/bmp`). Required for correct Volcengine Data URL formatting when `inputImage` is provided |
+| `inputImages` | array | - | Multiple input images for multi-image composition. Each item uses `{ data, mimeType }`; Volcengine converts them to `data:image/<format>;base64,<data>` entries in the `image` array |
+| `inputImagePaths` | array | - | Multiple input image file paths for multi-image composition. Supported by Gemini and by Volcengine when mapped to reference-image arrays |
 | `returnBase64` | boolean | - | Return the generated image as base64 data in the response. Image is always saved to disk regardless |
 | `fileName` | string | - | Custom filename for output (auto-generated if not specified). Extension is auto-appended based on output format if omitted |
 | `skipPromptEnhancement` | boolean | - | Skip prompt enhancement and use the prompt as-is. Recommended for multi-image blending. Overrides `SKIP_PROMPT_ENHANCEMENT` env var. Default: `false` |
-| `aspectRatio` | string | - | `1:1` (default), `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`, `1:4`, `1:8`, `4:1`, `8:1` |
-| `imageSize` | string | - | `1K`, `2K`, `4K`. Leave unspecified for standard quality |
+| `aspectRatio` | string | - | `1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`, `1:4`, `1:8`, `4:1`, `8:1`. When omitted, Gemini and Volcengine default to `16:9` |
+| `imageRequests` | array | - | Backward-compatible per-image prompts for `generate_image`. For new grouped multi-image requests, prefer `generate_multi_image` |
+| `imageSize` | string | - | `1K`, `2K`, `4K`. When omitted, Gemini and Volcengine default to `4K`. Volcengine further normalizes the final `WxH` into the provider legal pixel range |
 | `blendImages` | boolean | - | Enable multi-image blending for combining multiple visual elements naturally |
 | `maintainCharacterConsistency` | boolean | - | Maintain character appearance consistency across different poses and scenes |
 | `useWorldKnowledge` | boolean | - | Use real-world knowledge for accurate context (historical figures, landmarks, factual scenarios) |
 | `useGoogleSearch` | boolean | - | Enable Google Search grounding for real-time factual accuracy |
-| `purpose` | string | - | Intended use (e.g., "cookbook cover", "social media post"). Helps tailor visual style and details |
+| `purpose` | string | - | Intended use (e.g., "cookbook cover", "social media post"). Helps tailor visual style and details |
+
+### `generate_multi_image` Tool
+
+Use `generate_multi_image` for grouped multi-image generation in a single tool call. This is the preferred entry for Notebook planners that might otherwise split one user request into multiple `generate_image` calls.
+
+#### Recommended patterns
+
+- Use `outputCount` when the user wants multiple images with shared overall constraints.
+- Use `imageRequests` when the user wants several distinct images in one grouped request.
+- If `outputCount` is omitted, the server will try to infer it from phrases such as `4张图`, `四张海报`, or `4 images`.
+
+#### Example
+
+```json
+{
+  "prompt": "请生成4张统一风格的电商产品图，主题都是同一款极简白色保温杯，分别覆盖主图、侧面细节、手持使用场景、办公桌场景。要求返回4张独立图片，不要拆成多次生成。",
+  "provider": "volcengine"
+}
+```
 
 #### Response
 

package/dist/api/geminiClient.d.ts

@@ -3,22 +3,10 @@
  * Integrates with Google's Gemini AI API using the official SDK
  * Supports automatic URL Context processing and feature parameters
  */
-import type { ImageQuality } from '../types/mcp.js';
+import type { GeneratedImageResult, ImageProviderClient } from './imageProvider.js';
 import type { Result } from '../types/result.js';
 import type { Config } from '../utils/config.js';
 import { GeminiAPIError, NetworkError } from '../utils/errors.js';
-/**
- * Metadata for generated images
- */
-export interface GeminiGenerationMetadata {
-    model: string;
-    prompt: string;
-    mimeType: string;
-    timestamp: Date;
-    inputImageProvided: boolean;
-    modelVersion?: string;
-    responseId?: string;
-}
 /**
  * Parameters for Gemini API image generation
  */
@@ -33,25 +21,10 @@ export interface GeminiApiParams {
     aspectRatio?: string;
     imageSize?: string;
     useGoogleSearch?: boolean;
-    quality?: ImageQuality;
+    quality?: 'fast' | 'balanced' | 'quality';
 }
-/**
- * Result of image generation
- */
-export interface GeneratedImageResult {
-    imageData: Buffer;
-    metadata: GeminiGenerationMetadata;
-}
-/**
- * Gemini API client interface
- */
-export interface GeminiClient {
+export interface GeminiClient extends ImageProviderClient {
     generateImage(params: GeminiApiParams): Promise<Result<GeneratedImageResult, GeminiAPIError | NetworkError>>;
 }
-/**
- * Creates a new Gemini API client
- * @param config Configuration containing API key and other settings
- * @returns Result containing the client or an error
- */
 export declare function createGeminiClient(config: Config): Result<GeminiClient, GeminiAPIError>;
 //# sourceMappingURL=geminiClient.d.ts.map

package/dist/api/geminiClient.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"geminiClient.d.ts","sourceRoot":"","sources":["../../src/api/geminiClient.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAGH,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAA;AAEnD,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,oBAAoB,CAAA;AAEhD,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,oBAAoB,CAAA;AAChD,OAAO,EAAE,cAAc,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAA;AAuHjE;;GAEG;AACH,MAAM,WAAW,wBAAwB;IACvC,KAAK,EAAE,MAAM,CAAA;IACb,MAAM,EAAE,MAAM,CAAA;IACd,QAAQ,EAAE,MAAM,CAAA;IAChB,SAAS,EAAE,IAAI,CAAA;IACf,kBAAkB,EAAE,OAAO,CAAA;IAE3B,YAAY,CAAC,EAAE,MAAM,CAAA;IACrB,UAAU,CAAC,EAAE,MAAM,CAAA;CACpB;AAED;;GAEG;AACH,MAAM,WAAW,eAAe;IAC9B,MAAM,EAAE,MAAM,CAAA;IACd,UAAU,CAAC,EAAE,MAAM,CAAA;IACnB,kBAAkB,CAAC,EAAE,MAAM,CAAA;IAC3B,WAAW,CAAC,EAAE,KAAK,CAAC;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,QAAQ,EAAE,MAAM,CAAA;KAAE,CAAC,CAAA;IACvD,WAAW,CAAC,EAAE,MAAM,CAAA;IACpB,SAAS,CAAC,EAAE,MAAM,CAAA;IAClB,eAAe,CAAC,EAAE,OAAO,CAAA;IACzB,OAAO,CAAC,EAAE,YAAY,CAAA;CACvB;AAED;;GAEG;AACH,MAAM,WAAW,oBAAoB;IACnC,SAAS,EAAE,MAAM,CAAA;IACjB,QAAQ,EAAE,wBAAwB,CAAA;CACnC;AAED;;GAEG;AACH,MAAM,WAAW,YAAY;IAC3B,aAAa,CACX,MAAM,EAAE,eAAe,GACtB,OAAO,CAAC,MAAM,CAAC,oBAAoB,EAAE,cAAc,GAAG,YAAY,CAAC,CAAC,CAAA;CACxE;AA6VD;;;;GAIG;AACH,wBAAgB,kBAAkB,CAAC,MAAM,EAAE,MAAM,GAAG,MAAM,CAAC,YAAY,EAAE,cAAc,CAAC,CAmDvF"}
+{"version":3,"file":"geminiClient.d.ts","sourceRoot":"","sources":["../../src/api/geminiClient.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAGH,OAAO,KAAK,EAA0B,oBAAoB,EAAE,mBAAmB,EAAE,MAAM,oBAAoB,CAAA;AAE3G,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,oBAAoB,CAAA;AAEhD,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,oBAAoB,CAAA;AAChD,OAAO,EAAE,cAAc,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAA;AAkHjE;;GAEG;AACH,MAAM,WAAW,eAAe;IAC9B,MAAM,EAAE,MAAM,CAAA;IACd,UAAU,CAAC,EAAE,MAAM,CAAA;IACnB,kBAAkB,CAAC,EAAE,MAAM,CAAA;IAC3B,WAAW,CAAC,EAAE,KAAK,CAAC;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,QAAQ,EAAE,MAAM,CAAA;KAAE,CAAC,CAAA;IACvD,WAAW,CAAC,EAAE,MAAM,CAAA;IACpB,SAAS,CAAC,EAAE,MAAM,CAAA;IAClB,eAAe,CAAC,EAAE,OAAO,CAAA;IACzB,OAAO,CAAC,EAAE,MAAM,GAAG,UAAU,GAAG,SAAS,CAAA;CAC1C;AAED,MAAM,WAAW,YAAa,SAAQ,mBAAmB;IACvD,aAAa,CACX,MAAM,EAAE,eAAe,GACtB,OAAO,CAAC,MAAM,CAAC,oBAAoB,EAAE,cAAc,GAAG,YAAY,CAAC,CAAC,CAAA;CACxE;AA6TD,wBAAgB,kBAAkB,CAAC,MAAM,EAAE,MAAM,GAAG,MAAM,CAAC,YAAY,EAAE,cAAc,CAAC,CAiDvF"}