### Response Example for Download Image (Search + Compression)

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Example of a successful response after downloading an image using search mode and applying compression. It includes details about the original and compressed sizes, compression ratio, and quality settings.

```json
{
  "query": "mountain sunset",
  "imageId": 12345,
  "width": 1920,
  "height": 1080,
  "imageSize": 0,
  "user": "John Doe",
  "tags": "mountain sunset",
  "inputPath": "...",
  "outputPath": "./images/mountain.webp",
  "originalSize": 500000,
  "compressedSize": 80000,
  "compressionRatio": 84.00,
  "success": true,
  "iterations": 1,
  "qualitySteps": [85],
  "finalQuality": 85,
  "exceededTarget": false
}
```

--------------------------------

### Example Usage of Get Acknowledgement Tool

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/06-get-acknowledgement.md

Demonstrates how to call the get_acknowledgement tool with an empty arguments object, as no parameters are required.

```json
{
  "tool": "get_acknowledgement",
  "arguments": {}
}
```

--------------------------------

### Install image-processor-mcp via npm

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/07-credits-installation.md

Use this command to install the package from npm. Ensure you have Node.js 18 or higher and npm installed.

```bash
npm install image-processor-mcp
```

--------------------------------

### Sync Filenames Dry Run Example Output

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Example output demonstrating a dry run of the sync_filenames function. It indicates that no files were modified and provides a summary of the operation, including the mapping source, files scanned, and replacements found.

```text
Dry run — no files were modified.

Mapping source: .image-processor-mcp/mappings/mappings-2026-05-29-15-30-00.json
Files scanned: 42
Files with matches: 3
Total replacements: 12

--- Matches ---

[src/components/Header.tsx]
  "logo.png" → "logo.webp" (5x)
  Preview: ... context lines ...
  Line 15:23  "logo.png" → "logo.webp"
  ... and 3 more (see report)

Report saved to: .image-processor-mcp/syncs/dry-run/sync-report-2026-05-29-15-31-00.md
```

--------------------------------

### Sync Report Example

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/04-sync-filenames.md

Illustrates the structure and content of a generated sync report, including dry run status, file statistics, and detailed match information.

```text
Dry run — no files were modified.

Mapping source: .image-processor-mcp/mappings/mappings-2026-05-29-15-30-00.json
Files scanned: 42
Files with matches: 3
Total replacements: 12

--- Matches ---

[src/components/Header.tsx]
  "logo.png" → "logo.webp" (5x)
  Preview: ... context lines ...
  Line 15:23  "logo.png" → "logo.webp"
  Line 42:8  "logo.png" → "logo.webp"
  ... and 3 more (see report)

Report saved to: .image-processor-mcp/syncs/dry-run/sync-report-2026-05-29-15-31-00.md
```

--------------------------------

### JSON Response Example (with Rectangle and Blocks)

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

Response including rectangle details and detailed text blocks when specified in the request.

```json
{
  "success": true,
  "text": "Region-specific text",
  "confidence": 88.3,
  "words": 12,
  "language": "eng",
  "durationMs": 890,
  "preprocessed": true,
  "rectangle": { "top": 50, "left": 0, "width": 400, "height": 150 },
  "blocks": [ ... ]
}
```

--------------------------------

### Internal Implementation of Get Acknowledgement

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/06-get-acknowledgement.md

Illustrates the synchronous process of building the markdown documentation string by interpolating values from various configuration sources.

```typescript
ToolController.handleGetAcknowledgement()
  └─ Synchronous — no abort signal, no async operations
  └─ Builds markdown string via template literals
  └─ Interpolates values from:
      ├─ DEFAULTS (all default values)
      ├─ CONFIG_DESCRIPTIONS (parameter descriptions)
      ├─ SUPPORTED_EXTENSIONS (input format list)
      ├─ INPUT_EXTENSION_DESCRIPTIONS (format descriptions)
      ├─ SUPPORTED_OUTPUT_FORMATS (output format list)
      ├─ OUTPUT_FORMAT_DESCRIPTIONS (output format descriptions)
      ├─ RESTRICTED_EXTENSIONS (SVG)
      ├─ DEFAULT_EXTENSIONS (sync scan extensions)
      ├─ DEFAULT_EXCLUDE_DIRS (sync scan exclude dirs)
      └─ APP_CONFIG (name, version)
```

--------------------------------

### Run image-processor-mcp Directly

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/07-credits-installation.md

Execute this command to start the MCP server directly. It will run on stdio, outputting diagnostic messages to stderr and communicating with the MCP client via stdin/stdout.

```bash
npx -y image-processor-mcp
```

--------------------------------

### Response Example for compress_directory Tool

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/03-compress-directory.md

This JSON object represents a successful response from the compress_directory tool, detailing the outcome of the compression process, including file counts, sizes, and paths to generated reports.

```json
{
  "success": true,
  "totalFiles": 10,
  "successful": 9,
  "failed": 1,
  "totalOriginalSize": 5000000,
  "totalCompressedSize": 1200000,
  "overallReduction": 76.00,
  "durationMs": 5432,
  "reportPath": ".image-processor-mcp/reports/scan-report-2026-05-29-15-30-00.md",
  "mappingsPath": ".image-processor-mcp/mappings/mappings-2026-05-29-15-30-00.json"
}
```

--------------------------------

### JSON Response Example (Basic)

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

A typical JSON response when extracting text with default settings.

```json
{
  "success": true,
  "text": "Extracted text content here...",
  "confidence": 92.5,
  "words": 42,
  "language": "eng",
  "durationMs": 1234,
  "preprocessed": true
}
```

--------------------------------

### Response Example for Image Compression

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/02-compress-image.md

Illustrates the structure of the JSON response returned after an image compression operation, including details on original and compressed sizes, compression ratio, and success status.

```json
{
  "inputPath": "./input/photo.jpg",
  "outputPath": "./output/photo.webp",
  "originalSize": 500000,
  "compressedSize": 80000,
  "compressionRatio": 84.00,
  "success": true,
  "iterations": 1,
  "qualitySteps": [85],
  "finalQuality": 85,
  "exceededTarget": false
}
```

--------------------------------

### Get Acknowledgement Response Format

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/06-get-acknowledgement.md

Defines the structure of the response from the get_acknowledgement tool, which contains the generated markdown documentation.

```json
{
  "content": [{ "type": "text", "text": "<full markdown documentation>" }]
}
```

--------------------------------

### Extract Text with Default Parameters

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

Extracts text using default settings (English language, text output, preprocessing enabled). Provide the image path to start.

```javascript
import { extractText } from "@/modules/ocr/extract-text";

const result = await extractText({ imagePath: "/path/to/your/image.png" });
console.log(result.text);
```

--------------------------------

### Handle Get Acknowledgement in TypeScript

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

This function is used to retrieve comprehensive documentation of all tools, defaults, formats, and environment variables. It is synchronous and does not accept an abort signal or operate asynchronously.

```typescript
ToolController.handleGetAcknowledgement()
  // Synchronous — no abort signal, no async
  // Returns { content: [{ type: 'text', text: markdown }]
}
```

--------------------------------

### OCR Response Example (text format)

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

This JSON structure represents the successful extraction of text from an image. It includes the extracted text, confidence score, word count, language used, and processing duration.

```json
{
  "success": true,
  "text": "Extracted text content",
  "confidence": 92.5,
  "words": 42,
  "language": "eng",
  "durationMs": 1234,
  "preprocessed": true
}
```

--------------------------------

### Image Compression Algorithm Steps

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Outlines the sequential steps involved in the core compression algorithm, including reading metadata, resizing, and iterative quality reduction for recursive compression.

```text
1. Read original size + image metadata (sharp)
2. Compute resize:
   - Explicit w/h? → use those
   - Exceeds maxDimension? → scale proportionally
   - Otherwise → no resize
3. Loop:
   - sharp pipeline → apply format with quality
   - If input === output → buffer (safe in-place)
   - Else → write to file
   - If recursive + over target + above floor → reduce quality, loop
   - Else → break
4. Return CompressionResult
```

--------------------------------

### Output Path Algorithm

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Defines the algorithm for generating output paths, including directory normalization, file naming, and extension replacement.

```text
relativePath = inputDir → inputFile difference

1. Dir naming enabled → normalize each directory component
2. File naming enabled → normalize basename
3. Replace extension with .{outputFormat}
4. Prepend outputDir
```

--------------------------------

### Download Image Data Flow

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Illustrates the data flow for the `download_image` function, from request handling to image compression and final output.

```text
Request → type guard → Controller
  ├─ Search mode? → Pexels API → first hit URL
  └─ URL mode?    → use directly
  ↓
axios.get(url, arraybuffer, 30s timeout)
  ↓
Write to temp file → check compression params
  ├─ YES → ImageService.compressImage → clean temp → return result
  └─ NO  → move to final path → return success
```

--------------------------------

### Configure MCP Server for Claude Desktop

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/07-credits-installation.md

Add this JSON configuration to your `claude_desktop_config.json` to integrate the image-processor-mcp as an MCP server. The Pexels API key is optional and used for keyword-based image search.

```json
{
  "mcpServers": {
    "image-processor-mcp": {
      "command": "npx",
      "args": ["-y", "image-processor-mcp"],
      "env": {
        "IMAGE_SEARCH_API_KEY": "your_pexels_api_key_here"
      }
    }
  }
}
```

--------------------------------

### get_acknowledgement

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Retrieves comprehensive documentation of all tools, defaults, formats, and environment variables used by the image processor. This endpoint is synchronous and does not accept any parameters.

```APIDOC
## get_acknowledgement

### Description
Returns comprehensive documentation of all tools, defaults, formats, and environment variables.

### Method
GET

### Endpoint
/acknowledgement

### Parameters
**None** — empty input schema.

### Response
#### Success Response (200)
- **content** (array) - An array containing objects with `type` and `text` properties, where `text` is a markdown document detailing the system's configuration.

### Response Example
```json
{
  "content": [
    {
      "type": "text",
      "text": "# Image Processor Documentation\n\n## Tools\n... (details about tools)\n\n## Configuration Defaults\n| Parameter | Default | Description |\n|-----------|---------|-------------|\n| quality | `85` | Compression quality for lossy formats (1–100) |
| ... | ... | ... |
"
    }
  ]
}
```

### Implementation Notes
This endpoint is implemented via `ToolController.handleGetAcknowledgement()` and is synchronous.
```

--------------------------------

### Data Flow for Sync Filenames

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Illustrates the data flow from request to file modification or preview during the filename synchronization process. It details how mappings are read, files are scanned, and replacements are made.

```text
Request → type guard → Controller
  ├─ mappingFilePath?
  │   YES → ReferenceService.readMappingFile → validate against report
  │   NO  → use inline mappings
  ├─ dryRun = args.dryRun !== false
  ├─ ReferenceService.scanReferences:
  │   ├─ Collect files recursively (filtered by extension, excluding dirs)
  │   ├─ For each file:
  │   │   ├─ For each mapping:
  │   │   │   ├─ Regex-escape oldFileName, count occurrences
  │   │   │   ├─ Build line-by-line details
  │   │   │   ├─ Build preview (3 lines context)
  │   │   │   └─ If !dryRun: replace in content
  │   │   └─ If modified: write file
  │   └─ Return UpdateRefResult
  └─ ReportService.generateSyncReport → .image-processor-mcp/syncs/
```

--------------------------------

### Inline Mapping Configuration

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/04-sync-filenames.md

Use this configuration as a fallback when direct mapping objects are provided. Specify the root directory to scan and an array of `{oldFileName, newFileName}` objects.

```json
{
  "scanDir": "./src",
  "mappings": [
    { "oldFileName": "logo.png", "newFileName": "logo.webp" },
    { "oldFileName": "banner.jpg", "newFileName": "banner.avif" }
  ],
  "dryRun": true
}
```

--------------------------------

### Compress Image with Options

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/02-compress-image.md

Compresses an image to WebP format with specified quality and dimensions. Ensure the output directory exists before calling.

```typescript
import { compressImage } from "@/lib/image-processor-mcp/compress-image";

const inputPath = "./input/photo.jpg";
const outputPath = "./output/photo.webp";

await compressImage(inputPath, outputPath, {
  outputFormat: "webp",
  quality: 80,
  width: 500,
  height: 300,
});
```

--------------------------------

### Filename Normalization Logic

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Explains the process of normalizing filenames by replacing characters, collapsing underscores, trimming edges, and applying case conversion.

```text
Replace each char → '_', collapse multiple '_', trim edges, apply case
```

--------------------------------

### Data Flow for Image Compression

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Illustrates the data flow from request to file processing and report generation for image compression.

```text
Request → type guard → Controller
  ├─ Validate input directory exists
  ├─ Apply defaults, build FileNamingOptions
  ├─ Scan all image files (recursive)
  ├─ Filter out SVG (restricted)
  ├─ For each file:
  │   ├─ Compute output path (preserves structure, normalizes, changes ext)
  │   └─ Run ImageService.compressImage
  ├─ Generate report (.image-processor-mcp/reports/)
  └─ Generate mappings JSON (.image-processor-mcp/mappings/)
```

--------------------------------

### File-based Mapping Configuration

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/04-sync-filenames.md

Use this configuration when you have a `mappings-{timestamp}.json` file generated by `compress_directory`. Specify the root directory to scan and the path to the mapping file.

```json
{
  "scanDir": "./src",
  "mappingFilePath": ".image-processor-mcp/mappings/mappings-2026-05-29-15-30-00.json",
  "dryRun": true
}
```

--------------------------------

### Generated Files Structure

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Shows the directory structure for generated report and mappings files.

```text
.image-processor-mcp/
├── reports/scan-report-{timestamp}.md
└── mappings/mappings-{timestamp}.json
```

--------------------------------

### Batch Compress Directory

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Compresses all images in a directory recursively, preserving folder structure. Generates a report and mappings file.

```markdown
## 3. compress_directory

Batch compress all images in a directory recursively, preserving folder structure. Generates report + mappings.
```

--------------------------------

### Generated Files for Sync Reports

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Shows the file paths for generated synchronization reports, differentiating between live updates and dry-run previews. The timestamp in the filename is crucial for linking reports to specific mapping files.

```text
.image-processor-mcp/syncs/sync-report-{timestamp}.md           (live)
.image-processor-mcp/syncs/dry-run/sync-report-{timestamp}.md   (dry-run)
```

--------------------------------

### Image Download Response (Search + Compression)

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/01-download-image.md

This JSON structure represents the response when an image is found via search and then compressed. It includes details about the search query, image metadata, compression results, and success status.

```json
{
  "query": "mountain sunset",
  "imageId": 12345,
  "width": 1920,
  "height": 1080,
  "imageSize": 0,
  "user": "John Doe",
  "tags": "mountain sunset landscape",
  "inputPath": "...",
  "outputPath": "./images/mountain.webp",
  "originalSize": 500000,
  "compressedSize": 80000,
  "compressionRatio": 84.00,
  "success": true,
  "iterations": 1,
  "qualitySteps": [85],
  "finalQuality": 85,
  "exceededTarget": false
}
```

--------------------------------

### Image Download Response (URL Only, No Compression)

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/01-download-image.md

This JSON structure represents the response when an image is downloaded directly from a URL without any compression. It confirms the success of the operation and provides the output path and file size.

```json
{
  "success": true,
  "outputPath": "./images/photo.jpg",
  "fileSize": 500000
}
```

--------------------------------

### compress_image

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/README.md

Compress or convert a single image with full control over quality, dimensions, format, and recursive compression.

```APIDOC
## compress_image

### Description
Compress or convert a single image with full control over quality, dimensions, format, and recursive compression.

### Parameters
#### Path Parameters
- `inputPath` (string) - Required - Source image path
- `outputPath` (string) - Required - Destination path

#### Query Parameters
- `outputFormat` (string) - Optional - Default: `webp` - Output format
- `quality` (number) - Optional - Default: `85` - 1–100
- `lossless` (boolean) - Optional - Default: `false` - Lossless mode
- `effort` (number) - Optional - Default: `6` - CPU effort 0–6
- `width` (number) - Optional - Target width
- `height` (number) - Optional - Target height
- `maxDimension` (number) - Optional - Default: `2000` - Auto-resize threshold
- `recursiveCompress` (boolean) - Optional - Default: `false` - Iterative targeting
- `expectedSizeKB` (number) - Optional - Default: `100` - Target KB
- `qualityStepDown` (number) - Optional - Default: `10` - Per-iteration decrease
- `minimumQualityFloor` (number) - Optional - Default: `10` - Minimum quality

### Response Example
```json
{
  "inputPath": "./input/photo.jpg",
  "outputPath": "./output/photo.webp",
  "originalSize": 500000,
  "compressedSize": 80000,
  "compressionRatio": 84.00,
  "success": true,
  "iterations": 1,
  "qualitySteps": [85],
  "finalQuality": 85,
  "exceededTarget": false
}
```

### Edge Cases
- **In-place compression** (same path): Buffer-based to avoid corruption
- **Missing input file**: `success: false` with error
- **Unknown format**: Falls back to WebP
- **Recursion floor**: Stops when quality can't decrease further
- **Abort**: Returns partial result with `success: false`
```

--------------------------------

### Image Compression with Abort Signal

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/02-compress-image.md

Compresses an image and allows the operation to be cancelled using an AbortSignal. If cancelled mid-recursion, it returns a partial result with an error message.

```typescript
import { compressImage } from "@/lib/image-processor-mcp/compress-image";

const controller = new AbortController();
const signal = controller.signal;

compressImage(
  "./input/very_large.tiff",
  "./output/very_large_compressed.avif",
  {
    lossless: true,
    signal: signal, // Pass the abort signal
  }
).catch((error) => {
  if (error.name === 'AbortError') {
    console.log('Image compression was cancelled.');
  } else {
    console.error('An error occurred:', error);
  }
});

// To cancel the operation:
// controller.abort();
```

--------------------------------

### In-place Image Compression

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/02-compress-image.md

Compresses an image and saves it to the same path, overwriting the original. This is handled safely using an internal buffer to prevent data corruption.

```typescript
import { compressImage } from "@/lib/image-processor-mcp/compress-image";

await compressImage(
  "./input/photo.jpg",
  "./input/photo.jpg", // Same input and output path
  {
    quality: 75,
    outputFormat: "jpeg",
  }
);
```

--------------------------------

### Recursive Image Compression

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/02-compress-image.md

Recursively compresses an image, reducing quality iteratively until it meets the expected size in KB. The minimum quality floor prevents excessive degradation.

```typescript
import { compressImage } from "@/lib/image-processor-mcp/compress-image";

await compressImage(
  "./input/large.png",
  "./output/large_compressed.png",
  {
    recursiveCompress: true,
    expectedSizeKB: 50,
    qualityStepDown: 5,
    minimumQualityFloor: 20,
  }
);
```

--------------------------------

### Extract Text with Specific Language and Output Format

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

Specify a language (e.g., French) and an output format (e.g., JSON blocks with positions).

```javascript
import { extractText } from "@/modules/ocr/extract-text";

const result = await extractText({
  imagePath: "/path/to/your/image.png",
  language: "fra",
  outputFormat: "blocks"
});
console.log(result.blocks);
```

--------------------------------

### Extract Text with Preprocessing Disabled

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

Disable automatic image preprocessing (grayscale, auto-rotate, resolution normalization) if you intend to handle these steps manually.

```javascript
import { extractText } from "@/modules/ocr/extract-text";

const result = await extractText({
  imagePath: "/path/to/your/image.png",
  preprocess: false
});
console.log(result.text);
```

--------------------------------

### Extract Text with Multiple Languages

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

Process an image containing text in multiple languages by combining language codes with a '+'.

```javascript
import { extractText } from "@/modules/ocr/extract-text";

const result = await extractText({
  imagePath: "/path/to/your/image.png",
  language: "eng+chi_sim"
});
console.log(result.text);
```

--------------------------------

### Generated Mappings JSON Structure

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/03-compress-directory.md

This JSON structure is generated when at least one successful compression occurs and a timestamp can be extracted from the report filename. It maps old filenames to new filenames.

```json
{
  "generatedAt": "2026-05-29T15:30:00.000Z",
  "tool": "compress_directory",
  "mappings": [
    { "oldFileName": "photo.png", "newFileName": "photo.webp" },
    { "oldFileName": "banner.jpg", "newFileName": "banner.avif" }
  ]
}
```

--------------------------------

### Extract Text with TSV Output

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

Obtain OCR results in a Tab-Separated Values (TSV) format, suitable for tabular data processing.

```javascript
import { extractText } from "@/modules/ocr/extract-text";

const result = await extractText({
  imagePath: "/path/to/your/image.png",
  outputFormat: "tsv"
});
console.log(result.tsv);
```

--------------------------------

### Extract Text with HOCR Output

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

Generate OCR results in an HTML-like structured markup format (hOCR) for detailed text and layout information.

```javascript
import { extractText } from "@/modules/ocr/extract-text";

const result = await extractText({
  imagePath: "/path/to/your/image.png",
  outputFormat: "hocr"
});
console.log(result.hocr);
```

--------------------------------

### Extract Text from a Specific Region

Source: https://github.com/pansuriyadhvanil/image-processor-mcp/blob/main/docs/05-extract-text.md

Limit OCR processing to a defined rectangular area within the image. Coordinates are relative to the image's top-left corner.

```javascript
import { extractText } from "@/modules/ocr/extract-text";

const result = await extractText({
  imagePath: "/path/to/your/image.png",
  rectangle: { top: 50, left: 0, width: 400, height: 150 }
});
console.log(result.text);
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.