@lutery/vision-mcp 1.0.0 → 1.0.1

package/README.md

@@ -1,428 +1,227 @@
 # Vision MCP
 
-MCP Server providing vision capabilities for LLMs via GLM-4.6V, SiliconFlow, and ModelScope. This server enables LLMs without native vision support or with expensive vision models to access cost-effective visual analysis capabilities.
+一个基于 STDIO 的 MCP Server，为不具备视觉能力（或视觉模型成本较高）的 LLM 提供统一的图片分析能力。通过切换 Provider（环境变量配置），即可使用不同平台/厂商的多模态模型。
 
-## Features
+## 支持的模型 / Provider
 
-- 🤖 **Multiple Model Support**: GLM-4.6V, SiliconFlow, and ModelScope vision models
-- 🖼️ **Flexible Image Input**: URL, base64 data URL, or local file paths
-- 📊 **Multiple Analysis Types**: Image description, UI analysis, object detection, OCR, and structured extraction
-- 🔧 **System Prompt Templates**: Built-in templates for common vision tasks
-- 📦 **Easy Deployment**: STDIO MCP Server, runs with npx
-- 🔒 **Secure**: Environment-based configuration, sensitive data masking in logs
+通过 `VISION_MODEL_TYPE` 选择提供商：
 
-### Streaming Response Support
+| type | Provider | 默认 `VISION_API_BASE_URL` | 默认 `VISION_MODEL_NAME` | 备注 |
+|------|----------|----------------------------|--------------------------|------|
+| `glm-4.6v` | 智谱 GLM-4.6V | `https://open.bigmodel.cn/api/paas/v4` | `glm-4.6v` | `glm` 是别名（等同 `glm-4.6v`） |
+| `glm` | GLM-4.6V（别名） | `https://open.bigmodel.cn/api/paas/v4` | `glm-4.6v` | 兼容历史配置 |
+| `siliconflow` | SiliconFlow（OpenAI 兼容） | `https://api.siliconflow.cn/v1` | `Qwen/Qwen2-VL-72B-Instruct` | 视觉模型丰富 |
+| `modelscope` | ModelScope API-Inference（OpenAI 兼容） | `https://api-inference.modelscope.cn/v1` | `ZhipuAI/GLM-4.6V` | 需实名/绑定阿里云，受限额影响 |
+| `openai` | OpenAI | `https://api.openai.com/v1` | `gpt-4o` | 适配 Chat Completions |
+| `claude` | Anthropic Claude（Messages API） | `https://api.anthropic.com` | `claude-3-5-sonnet-20241022` | `baseUrl` 不要带 `/v1` |
+| `gemini` | Google Gemini（generateContent API） | `https://api.gptsapi.net` | `gemini-2.0-flash-exp` | 默认是代理地址，可改为官方或自建网关 |
 
-Current adapters explicitly disable streaming responses (`stream: false`) and are designed for complete JSON responses. This ensures compatibility with both GLM-4.6V and SiliconFlow APIs.
+获取 API Key / Token（各平台控制台）：
 
-**Note**: Streaming-only providers are not currently supported. If a provider only supports streaming responses (Server-Sent Events/text/event-stream format), the adapter will fail as it expects a complete JSON response. To add support for streaming providers, a streaming response parser would need to be implemented.
+- GLM（智谱）：https://open.bigmodel.cn/
+- SiliconFlow：https://cloud.siliconflow.cn/
+- ModelScope：https://modelscope.cn/my/myaccesstoken
+- OpenAI：https://platform.openai.com/
+- Claude（Anthropic）：https://console.anthropic.com/
+- Gemini（Google AI）：https://ai.google.dev/
 
-## Quick Start
+## 特性
 
-### Installation
+- 多 Provider 一键切换（仅需改环境变量）
+- 图片输入：URL / base64 data URL / 本地文件路径
+- 内置系统提示词模板：UI 分析、OCR、目标检测、结构化提取等
+- 安全：日志自动脱敏 API Key，且会过滤模型返回的 thinking/reasoning 内容
+- 严格遵守 MCP：stdout 仅用于 JSON-RPC，日志走 stderr
 
-1. Clone or download this repository
-2. Install dependencies:
+## 安装与运行
 
-```bash
-cd vision_mcp
-npm install
-```
-
-### Configuration
+要求：Node.js >= 18
 
-Create a `.env` file in the project root:
+### 作为 NPM 包被 MCP 客户端启动（推荐）
 
-#### Option 1: GLM-4.6V
+在 MCP 客户端（如 Claude Desktop）里配置命令为 `npx`：
 
-```bash
-VISION_MODEL_TYPE=glm-4.6v
-VISION_MODEL_NAME=glm-4.6v
-VISION_API_BASE_URL=https://open.bigmodel.cn/api/paas/v4
-VISION_API_KEY=your-glm-api-key
+```json
+{
+  "mcpServers": {
+    "vision-mcp": {
+      "command": "npx",
+      "args": ["-y", "@lutery/vision-mcp"],
+      "env": {
+        "VISION_MODEL_TYPE": "siliconflow",
+        "VISION_API_KEY": "sk-your-key",
+        "VISION_MODEL_NAME": "Qwen/Qwen2-VL-72B-Instruct",
+        "VISION_API_BASE_URL": "https://api.siliconflow.cn/v1"
+      }
+    }
+  }
+}
 ```
 
-#### Option 2: SiliconFlow
-
-```bash
-VISION_MODEL_TYPE=siliconflow
-VISION_MODEL_NAME=Qwen/Qwen2-VL-72B-Instruct
-VISION_API_BASE_URL=https://api.siliconflow.cn/v1
-VISION_API_KEY=your-siliconflow-api-key
-```
+说明：
+- `VISION_MODEL_NAME` / `VISION_API_BASE_URL` 可省略（会使用该 provider 的默认值）
+- 如需更详细的配置项，建议直接参考 `.env.example`
 
-#### Option 3: ModelScope API-Inference
+也可以全局安装后直接使用可执行文件（`bin` 名称为 `vision-mcp`）：
 
 ```bash
-VISION_MODEL_TYPE=modelscope
-VISION_MODEL_NAME=ZhipuAI/GLM-4.6V
-VISION_API_BASE_URL=https://api-inference.modelscope.cn/v1
-VISION_API_KEY=your-modelscope-token
+npm i -g @lutery/vision-mcp
+vision-mcp
 ```
 
-**Note**: ModelScope requires:
-- Real-name authentication on your ModelScope account
-- Aliyun account binding
-- API usage limits apply (see [API Limits](https://www.modelscope.cn/docs/model-service/API-Inference/limits))
-
-### Build
+### 本地开发运行
 
 ```bash
+cd mcp/vision_mcp
+npm install
 npm run build
-```
-
-### Run (local)
-
-```bash
 node dist/index.js
 ```
 
-If successful, you'll see: `Vision MCP Server is running on stdio` in stderr.
-
-### Run (npx)
-
-```bash
-# Local package (requires build first)
-npx .
-
-# Published package
-npx -y @lutery/vision-mcp
-```
-
-## MCP Client Configuration
+成功启动后，会在 stderr 输出 `Vision MCP Server is running on stdio`。
 
-### Claude Desktop
+## 配置（环境变量）
 
-Add to your Claude Desktop configuration:
+最小必填：
 
-```json
-{
-  "mcpServers": {
-    "vision-mcp": {
-      "command": "npx",
-      "args": ["-y", "@lutery/vision-mcp"],
-      "env": {
-        "VISION_MODEL_TYPE": "glm-4.6v",
-        "VISION_MODEL_NAME": "glm-4.6v",
-        "VISION_API_BASE_URL": "https://open.bigmodel.cn/api/paas/v4",
-        "VISION_API_KEY": "your-api-key"
-      }
-    }
-  }
-}
-```
+- `VISION_MODEL_TYPE`：选择 provider
+- `VISION_API_KEY`：对应 provider 的 key/token
 
-Or with a local installation:
+常用可选：
 
-```json
-{
-  "mcpServers": {
-    "vision-mcp": {
-      "command": "node",
-      "args": ["/path/to/vision_mcp/dist/index.js"],
-      "env": {
-        "VISION_MODEL_TYPE": "glm-4.6v",
-        "VISION_MODEL_NAME": "glm-4.6v",
-        "VISION_API_BASE_URL": "https://open.bigmodel.cn/api/paas/v4",
-        "VISION_API_KEY": "your-api-key"
-      }
-    }
-  }
-}
-```
+| 变量 | 说明 | 默认 |
+|------|------|------|
+| `VISION_MODEL_NAME` | 模型名称 | 各 provider 内置默认值 |
+| `VISION_API_BASE_URL` | API 基础地址（不要带具体 endpoint） | 各 provider 内置默认值 |
+| `VISION_API_TIMEOUT` | 超时时间（毫秒） | `60000` |
+| `VISION_MAX_RETRIES` | 最大重试次数 | `2` |
+| `VISION_STRICT_URL_VALIDATION` | 严格校验图片 URL 是否以 `.jpg/.jpeg/.png/.webp` 结尾 | `true` |
+| `LOG_LEVEL` | 日志级别：`debug`/`info`/`warn`/`error` | `info` |
 
-### Cursor/Codex CLI
+Provider 特有配置：
 
-Similar configuration for other MCP-compatible clients.
+- Claude
+  - `VISION_CLAUDE_API_VERSION`：Anthropic API 版本（默认 `2023-06-01`）
+- Gemini
+  - `VISION_GEMINI_API_VERSION`：`v1beta` / `v1`（默认 `v1beta`）
+  - `VISION_GEMINI_AUTH_MODE`：`bearer` / `x-goog` / `query`（默认 `bearer`）
+  - `VISION_GEMINI_IMAGE_PART_MODE`：`inline_data` / `inline_bytes`（默认 `inline_data`）
 
-## Using the Tools
+## MCP 工具（Tools）
 
-### 1. Analyze Image
+本服务注册了 3 个工具：
 
-Main tool for image analysis:
+### 1) `analyze_image`
 
-```javascript
-// Tool: analyze_image
-// Parameters:
-{
-  "image": "https://example.com/image.jpg",        // Image URL, base64, or local path
-  "prompt": "Describe this UI design in detail",   // Analysis prompt
-  "output_format": "text",                          // Optional: "text" or "json"
-  "template": "ui-analysis"                         // Optional: see templates below
-}
-```
-
-#### Example Prompts
+参数：
 
-**UI Analysis:**
 ```json
 {
-  "image": "./screenshot.png",
-  "prompt": "Analyze this UI design and extract all UI components with their positions and styles",
+  "image": "https://example.com/a.png",
+  "prompt": "请描述这个界面有哪些组件",
+  "output_format": "text",
   "template": "ui-analysis"
 }
 ```
 
-**Object Detection:**
-```json
-{
-  "image": "https://example.com/photo.jpg",
-  "prompt": "Detect all objects and provide their coordinates",
-  "template": "object-detection"
-}
-```
+字段说明：
+- `image`：支持 URL / base64 data URL / 本地路径
+- `prompt`：你的分析任务描述
+- `output_format`：`text` 或 `json`（提示偏好；不会强制校验 JSON）
+- `template`：可选系统模板（见下方 `list_templates`）
 
-**OCR:**
-```json
-{
-  "image": "data:image/png;base64,iVBORw0KGgo...",
-  "prompt": "Extract all text from this image",
-  "template": "ocr"
-}
-```
+### 2) `list_templates`
 
-**Structured Extraction:**
-```json
-{
-  "image": "./form.jpg",
-  "prompt": "Extract all form fields and values as JSON",
-  "output_format": "json"
-}
-```
+列出内置系统提示词模板（包含 id、用途说明等）。
 
-### 2. List Templates
+### 3) `get_config`
 
-List available system prompt templates:
+返回当前生效的模型配置（API Key 会脱敏）。
 
-```javascript
-// Tool: list_templates
-// Parameters: none
-```
+## 图片输入规范
 
-Available templates:
-- `general-description` - General image description
-- `ui-analysis` - UI prototype and interface analysis
-- `object-detection` - Object detection and localization
-- `ocr` - Text extraction (OCR)
-- `structured-extraction` - Structured data extraction
+支持三种输入：
 
-### 3. Get Config
+1) URL
 
-Get current model configuration:
-
-```javascript
-// Tool: get_config
-// Parameters: none
+```text
+https://example.com/image.png
 ```
 
-## Image Input Formats
+默认开启严格校验：URL 必须以 `.jpg/.jpeg/.png/.webp` 结尾，否则报错。可通过 `VISION_STRICT_URL_VALIDATION=false` 放宽（仅告警）。
 
-### 1. URL
+2) Base64 Data URL
 
-```
-https://example.com/image.jpg
+```text
+data:image/png;base64,iVBORw0KGgo...
 ```
 
-### 2. Base64 Data URL
+支持的 MIME：`image/jpeg` / `image/jpg` / `image/png` / `image/webp`。
 
-```
-data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD...
-```
-
-### 3. Local File Path
+3) 本地文件路径
 
-```
-/path/to/image.png
-./relative/path/image.jpg
+```text
+./test/image.png
+D:\\path\\to\\image.jpg
 ```
 
-Note: Local paths only work if the MCP server has access to the filesystem.
-Note: URL validation is strict by default (see `VISION_STRICT_URL_VALIDATION`).
+要求 MCP Server 进程对该路径可读；仅支持 `.jpg/.jpeg/.png/.webp`。
 
-## Environment Variables
+补充：Gemini provider 不支持直接传 URL 图片，本项目会在 Gemini 适配器内下载 URL 并转 base64（有大小与超时限制）。
 
-| Variable | Description | Default | Required |
-|----------|-------------|---------|----------|
-| `VISION_MODEL_TYPE` | Model type: `glm` (alias for `glm-4.6v`), `glm-4.6v`, `siliconflow`, or `modelscope` | - | Yes |
-| `VISION_MODEL_NAME` | Model name for the API | See defaults below | Yes |
-| `VISION_API_BASE_URL` | API base URL (must be base path, no `/chat/completions`) | See defaults below | Yes |
-| `VISION_API_KEY` | API key for authentication | - | Yes |
-| `VISION_API_TIMEOUT` | Request timeout in milliseconds | 60000 | No |
-| `VISION_MAX_RETRIES` | Maximum retry attempts | 2 | No |
-| `VISION_STRICT_URL_VALIDATION` | Enforce strict image URL validation | `true` | No |
-| `LOG_LEVEL` | Log level: `debug`, `info`, `warn`, `error` | `info` | No |
+## 关于流式响应（Streaming）
 
-**Notes**:
-- `VISION_STRICT_URL_VALIDATION` defaults to `true`, enforcing strict validation that URLs must end with supported image extensions (`.jpg`, `.jpeg`, `.png`, `.webp`). Set to `false` to allow non-image URLs with a warning only.
-- For GLM-4.6V provider, both `glm` and `glm-4.6v` values work for `VISION_MODEL_TYPE`. `glm` is provided as a convenient alias.
+所有适配器均强制 `stream: false`，并按“完整 JSON 响应”进行解析。
 
-### Model Defaults
+如果某个上游只支持 SSE / `text/event-stream`，目前不支持（需要额外实现流式解析器）。
 
-**GLM-4.6V:**
-```bash
-VISION_MODEL_NAME=glm-4.6v
-VISION_API_BASE_URL=https://open.bigmodel.cn/api/paas/v4
-```
+## 开发与测试
 
-**SiliconFlow:**
 ```bash
-VISION_MODEL_NAME=Qwen/Qwen2-VL-72B-Instruct
-VISION_API_BASE_URL=https://api.siliconflow.cn/v1
+cd mcp/vision_mcp
+npm install
+npm run build
 ```
 
-## API Keys
-
-### GLM-4.6V
-
-Get your API key from: [智谱 AI 开放平台](https://open.bigmodel.cn/)
-
-Format: `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.xxxxxxxxxxxxxxxxxxxx`
-
-### SiliconFlow
-
-Get your API key from: [SiliconFlow](https://cloud.siliconflow.cn/)
-
-Format: `sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx`
-
-## MCP Protocol Note
+测试：
 
-**IMPORTANT**: This is a STDIO-based MCP Server. According to MCP protocol:
+- 仅跑单测（不需要任何 API Key）：
 
-- **DO NOT** use `console.log()` or write to stdout
-- **USE ONLY** `console.error()` for logging (stderr)
-- stdout is reserved for JSON-RPC communication
-
-The server handles this automatically. If you fork this project, ensure you follow this rule.
-
-## Development
-
-### Project Structure
-
-```
-vision_mcp/
-├── src/
-│   ├── index.ts              # MCP Server entry point
-│   ├── config/
-│   │   └── model-config.ts   # Configuration management
-│   ├── tools/
-│   │   └── vision-tool.ts    # Vision analysis tool
-│   ├── adapters/
-│   │   ├── base-adapter.ts   # Base adapter class
-│   │   ├── glm-adapter.ts    # GLM-4.6V adapter
-│   │   └── siliconflow-adapter.ts  # SiliconFlow adapter
-│   ├── prompts/
-│   │   └── system.ts         # System prompt templates
-│   └── utils/
-│       ├── errors.ts         # Error handling
-│       ├── logger.ts         # Logging utilities
-│       └── image-input.ts    # Image input normalization
-├── package.json
-├── tsconfig.json
-└── README.md
+```bash
+npm run test:unit
 ```
 
-### Building
+- 跑集成测试（需要配置好 `VISION_*` 环境变量）：
 
 ```bash
-# Install dependencies
-npm install
-
-# Build TypeScript
-npm run build
-
-# Run tests
 npm test
 ```
 
-### Testing Notes
-
-- `npm test` uses `VISION_API_KEY` (default) or provider-specific keys in the test script:
-  - `SILICONFLOW_API_KEY`
-  - `GLM_API_KEY`
-- If no API key is set, the tests will exit with a clear error message.
-
-## Troubleshooting
-
-### 1. "Failed to load model configuration"
-
-- Check all required environment variables are set
-- Verify `VISION_MODEL_TYPE` is either `glm-4.6v` or `siliconflow`
-
-### 2. "API Key not found"
-
-- Set `VISION_API_KEY` in your environment
-- Verify the API key format matches the model requirements
-
-### 3. "Connection timeout"
-
-- Increase `VISION_API_TIMEOUT` value
-- Check network connectivity to the API endpoint
-- Verify API endpoint URL is correct
-
-### 4. "Invalid image URL"
-
-- Ensure URL is publicly accessible
-- Check URL format (http:// or https://)
-- Verify image format is supported
-
-### 5. "Permission denied reading file"
-
-- MCP server needs filesystem access for local files
-- Use absolute paths or ensure relative paths are accessible
-- Check file permissions
-
-### 6. "Invalid API endpoint" or "404 Not Found"
-
-- Ensure `VISION_API_BASE_URL` is the base path only, without `/chat/completions`
-- Correct: `https://api.siliconflow.cn/v1`
-- Incorrect: `https://api.siliconflow.cn/v1/chat/completions`
-- Check the error details for the full request URL to diagnose endpoint issues
-
-## Security Notes
-
-- API keys are loaded from environment variables, never hardcoded
-- API keys are masked in logs
-- Images are not persisted by default
-- MCP server should run in trusted environments only (no built-in auth)
-- **Thinking/Reasoning Content Filtering**: Model thinking/reasoning content is automatically filtered from responses to prevent exposing internal reasoning to MCP clients. This filtering is unconditional and applied to all supported models regardless of configuration.
-
-## Security Best Practices
-
-⚠️ **IMPORTANT**: Never commit API keys or credentials to the repository!
+## 常见问题（Troubleshooting）
 
-- **Use environment variables** for sensitive data (`.env` file)
-- **Keep local test credentials** in `.gitignore`'d files (e.g., `test_key.local.md`)
-- **Rotate keys immediately** if accidentally exposed or committed
-- **See** `doc/test_key.example.md` for test setup template
-- **Never** copy real API keys into documentation, code comments, or issue trackers
+### 1) 配置加载失败：`Missing VISION_MODEL_TYPE` / `Unsupported model type`
 
-**Key Protection Checklist**:
-- [ ] `.env` is in `.gitignore`
-- [ ] `.env.local` is in `.gitignore`
-- [ ] No real keys in `test_key.md` (use `test_key.example.md` instead)
-- [ ] No keys in documentation or comments
-- [ ] Review git history for accidental key commits (`git log --all --full-history -S --source --all -- "*secret*" "*key*" "*password*" "test_key.md"`)
+- 确认设置了 `VISION_MODEL_TYPE`
+- 可用值：`glm` / `glm-4.6v` / `siliconflow` / `modelscope` / `openai` / `claude` / `gemini`
 
-## License
+### 2) `Missing VISION_API_KEY`
 
-MIT
+- 确认 `VISION_API_KEY` 已设置（在 `.env` 或 MCP 客户端 `env` 里）
 
-## Contributing
+### 3) 404 / endpoint 错误
 
-1. Fork the repository
-2. Create a feature branch
-3. Make your changes
-4. Add tests
-5. Submit a pull request
+- `VISION_API_BASE_URL` 必须是“base”，不要带具体 endpoint
+  - OpenAI / SiliconFlow / ModelScope：会自动拼 `/chat/completions`
+  - Claude：会自动拼 `/v1/messages`（`baseUrl` 不要写成 `.../v1`）
+  - Gemini：会自动拼 `/{apiVersion}/models/{model}:generateContent`
 
-## Support
+### 4) 图片 URL 校验失败
 
-For issues and questions:
-- Open an issue on the repository
-- Check model documentation:
-  - [GLM-4.6V Docs](https://docs.bigmodel.cn/)
-  - [SiliconFlow Docs](https://docs.siliconflow.cn/)
+- 默认要求 URL 以 `.jpg/.jpeg/.png/.webp` 结尾
+- 如需放宽：`VISION_STRICT_URL_VALIDATION=false`
 
+## 安全说明
 
-## TODO
-- [ ] 适配modelscope的视觉模型接口请求：https://www.modelscope.cn/docs/model-service/API-Inference/intro
+- 不要在 stdout 打日志（stdout 仅用于 MCP JSON-RPC），本项目日志统一走 stderr
+- API Key 会在日志中脱敏
+- 会无条件过滤模型返回的 thinking/reasoning 内容，避免泄露内部推理信息

package/dist/adapters/gemini-adapter.d.ts

@@ -0,0 +1,73 @@
+/**
+ * Gemini Adapter
+ *
+ * @description Gemini generateContent API 适配器实现，支持 Google Gemini 多模态视觉模型
+ * @see https://ai.google.dev/api/rest/v1beta/models/generateContent
+ */
+import { BaseVisionModelAdapter, VisionModelResponse } from './base-adapter.js';
+import { ModelConfig } from '../config/model-config.js';
+export interface GeminiAdapterOptions {
+    apiVersion?: string;
+    authMode?: 'bearer' | 'x-goog' | 'query';
+    imagePartMode?: 'inline_data' | 'inline_bytes';
+    maxTokens?: number;
+}
+export declare class GeminiAdapter extends BaseVisionModelAdapter {
+    private options;
+    constructor(config: ModelConfig, options?: GeminiAdapterOptions);
+    analyze(imageData: string, prompt: string): Promise<string>;
+    analyzeWithResponse(imageData: string, prompt: string): Promise<VisionModelResponse>;
+    private callGeminiAPI;
+    /**
+     * 构建 API URL
+     * Format: {baseUrl}/{apiVersion}/models/{model}:generateContent
+     */
+    private buildApiUrl;
+    /**
+     * 构建认证头
+     */
+    private buildAuthHeaders;
+    /**
+     * 构建请求体
+     */
+    private buildRequest;
+    /**
+     * 构建图片 part
+     * 支持三种输入格式：
+     * 1. HTTP(S) URL - 下载并转换为 base64
+     * 2. Data URL - 直接解析
+     * 3. Base64 字符串 - 假设为图片
+     */
+    private buildImagePart;
+    /**
+     * 创建 inline_data part
+     */
+    private createInlineDataPart;
+    /**
+     * 脱敏 URL 用于日志记录（移除敏感的查询参数）
+     * @param url - 原始 URL
+     * @returns 脱敏后的 URL（query 模式下的 key 会被替换为 ***）
+     */
+    private sanitizeUrl;
+    /**
+     * 下载图片并转换为 base64
+     * Gemini 不支持直接的 URL 输入，必须先下载
+     */
+    private downloadImageAsBase64;
+    /**
+     * 处理错误响应
+     */
+    private handleErrorResponse;
+    /**
+     * 解析响应数据
+     */
+    private parseResponseData;
+    /**
+     * 归一化 Gemini 响应为 VisionModelResponse
+     * 支持两种响应格式：
+     * 1. 官方格式: candidates[0].content.parts[].text
+     * 2. 代理格式: candidates[0].output.parts[].text
+     */
+    private normalizeResponse;
+}
+//# sourceMappingURL=gemini-adapter.d.ts.map

package/dist/adapters/gemini-adapter.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"gemini-adapter.d.ts","sourceRoot":"","sources":["../../src/adapters/gemini-adapter.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,EAAE,sBAAsB,EAAE,mBAAmB,EAAE,MAAM,mBAAmB,CAAC;AAChF,OAAO,EAAE,WAAW,EAAE,MAAM,2BAA2B,CAAC;AAMxD,MAAM,WAAW,oBAAoB;IACnC,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,QAAQ,CAAC,EAAE,QAAQ,GAAG,QAAQ,GAAG,OAAO,CAAC;IACzC,aAAa,CAAC,EAAE,aAAa,GAAG,cAAc,CAAC;IAC/C,SAAS,CAAC,EAAE,MAAM,CAAC;CACpB;AAOD,qBAAa,aAAc,SAAQ,sBAAsB;IACvD,OAAO,CAAC,OAAO,CAAiC;gBAEpC,MAAM,EAAE,WAAW,EAAE,OAAO,GAAE,oBAAyB;IAgC7D,OAAO,CAAC,SAAS,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC;IAyB3D,mBAAmB,CAAC,SAAS,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,OAAO,CAAC,mBAAmB,CAAC;YA0B5E,aAAa;IAuD3B;;;OAGG;IACH,OAAO,CAAC,WAAW;IAcnB;;OAEG;IACH,OAAO,CAAC,gBAAgB;IAcxB;;OAEG;YACW,YAAY;IAmB1B;;;;;;OAMG;YACW,cAAc;IAmB5B;;OAEG;IACH,OAAO,CAAC,oBAAoB;IAiB5B;;;;OAIG;IACH,OAAO,CAAC,WAAW;IAOnB;;;OAGG;YACW,qBAAqB;IAuEnC;;OAEG;YACW,mBAAmB;IAgCjC;;OAEG;YACW,iBAAiB;IAiC/B;;;;;OAKG;IACH,OAAO,CAAC,iBAAiB;CAsE1B"}