diff --git a/.gitignore b/.gitignore
index 3c3629e..4f228ac 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1,9 @@
 node_modules
+**/dist/
+**/examples/
+**/*.traineddata
+.env
+.env.local
+tmp/
+!tmp/.gitkeep
+*.tsbuildinfo
\ No newline at end of file
diff --git a/cli/README.md b/cli/README.md
new file mode 100644
index 0000000..cb67c7c
--- /dev/null
+++ b/cli/README.md
@@ -0,0 +1,254 @@
+# Documind CLI
+
+Command-line interface for intelligent document processing and structured data extraction using Documind.
+
+## Installation
+
+```bash
+# Install dependencies
+npm install
+
+# Build the CLI
+npm run build
+```
+
+## Usage
+
+The CLI provides several commands to work with documents:
+
+### Extract Command
+
+Extract structured data from documents using schemas, templates, or auto-generation.
+
+```bash
+documind extract [options]
+```
+
+#### Options
+
+- `-f, --file <path>` - Path to the document file (required)
+- `-m, --model <model>` - LLM model to use (default: gpt-4o-mini)
+- `-s, --schema <path>` - Path to JSON schema file
+- `-t, --template <name>` - Name of a predefined template
+- `-a, --auto-schema` - Auto-generate schema from document
+- `-i, --instructions <text>` - Instructions for auto-schema generation
+- `-o, --output <path>` - Output file path for results (JSON)
+- `--base-url <url>` - Base URL for local LLM (for Ollama)
+
+#### Examples
+
+**Extract with auto-generated schema:**
+```bash
+documind extract -f invoice.pdf --auto-schema
+```
+
+**Extract with custom schema:**
+```bash
+documind extract -f document.pdf -s schema.json -o output.json
+```
+
+**Extract using a predefined template:**
+```bash
+documind extract -f invoice.pdf -t invoice -o result.json
+```
+
+**Extract with local LLM (Ollama):**
+```bash
+documind extract -f doc.pdf -m llama3.2-vision --base-url http://localhost:11434/v1 --auto-schema
+```
+
+**Extract with specific instructions:**
+```bash
+documind extract -f contract.pdf --auto-schema -i "Extract party names, dates, and monetary amounts"
+```
+
+### Convert Command
+
+Convert documents to markdown or plain text.
+
+```bash
+documind convert [options]
+```
+
+#### Options
+
+- `-f, --file <path>` - Path to the document file (required)
+- `-m, --model <model>` - LLM model to use (default: gpt-4o-mini)
+- `-t, --format <format>` - Output format: markdown or plaintext (default: markdown)
+- `-o, --output <path>` - Output file path
+- `--base-url <url>` - Base URL for local LLM (for Ollama)
+
+#### Examples
+
+**Convert to markdown:**
+```bash
+documind convert -f document.pdf -t markdown -o output.md
+```
+
+**Convert to plain text:**
+```bash
+documind convert -f document.pdf -t plaintext -o output.txt
+```
+
+### Templates Command
+
+List all available predefined templates.
+
+```bash
+documind templates
+```
+
+## Supported Models
+
+### OpenAI Models
+- `gpt-4o`
+- `gpt-4o-mini` (default)
+- `gpt-4.1`
+- `gpt-4.1-mini`
+
+### Google Models
+- `gemini-2.0-flash-001`
+- `gemini-2.0-flash-lite-preview-02-05`
+- `gemini-1.5-flash`
+- `gemini-1.5-flash-8b`
+- `gemini-1.5-pro`
+
+### Local Models (Ollama)
+- `llama3.2-vision`
+
+## Environment Variables
+
+Set these environment variables for API access:
+
+```bash
+# OpenAI API key for GPT models
+export OPENAI_API_KEY="your-openai-api-key"
+
+# Google Gemini API key
+export GEMINI_API_KEY="your-gemini-api-key"
+
+# Base URL for local LLM (Ollama)
+export BASE_URL="http://localhost:11434/v1"
+```
+
+Or create a `.env` file:
+
+```env
+OPENAI_API_KEY=your-openai-api-key
+GEMINI_API_KEY=your-gemini-api-key
+BASE_URL=http://localhost:11434/v1
+```
+
+## Schema Format
+
+Schemas should be in JSON format with the following structure:
+
+```json
+[
+  {
+    "name": "field_name",
+    "type": "string",
+    "description": "Field description"
+  },
+  {
+    "name": "nested_object",
+    "type": "object",
+    "description": "An object field",
+    "children": [
+      {
+        "name": "child_field",
+        "type": "number",
+        "description": "Child field description"
+      }
+    ]
+  },
+  {
+    "name": "items_list",
+    "type": "array",
+    "description": "A list of items",
+    "children": [
+      {
+        "name": "item_name",
+        "type": "string",
+        "description": "Item name"
+      }
+    ]
+  }
+]
+```
+
+### Supported Field Types
+
+- `string` - Text values
+- `number` - Numeric values
+- `boolean` - True/false values
+- `enum` - Predefined set of values (requires `values` array)
+- `object` - Nested objects (requires `children` array)
+- `array` - Lists (requires `children` array)
+
+## Using with Local LLMs
+
+To use Ollama or other local LLMs:
+
+1. Install and start Ollama:
+```bash
+# Install Ollama
+curl https://ollama.ai/install.sh | sh
+
+# Pull the vision model
+ollama pull llama3.2-vision
+
+# Start Ollama (if not already running)
+ollama serve
+```
+
+2. Use with the CLI:
+```bash
+documind extract \
+  -f document.pdf \
+  -m llama3.2-vision \
+  --base-url http://localhost:11434/v1 \
+  --auto-schema \
+  -o result.json
+```
+
+## Development
+
+```bash
+# Watch mode
+npm run dev
+
+# Build
+npm run build
+
+# Run
+npm start -- extract -f example.pdf --auto-schema
+```
+
+## Troubleshooting
+
+### "Cannot find module" errors
+
+Make sure you've built the project:
+```bash
+npm run build
+```
+
+### API Key errors
+
+Ensure your API keys are set:
+```bash
+echo $OPENAI_API_KEY
+echo $GEMINI_API_KEY
+```
+
+### Local LLM connection issues
+
+Check if Ollama is running:
+```bash
+curl http://localhost:11434/v1/models
+```
+
+## License
+
+AGPL-3.0
diff --git a/cli/package.json b/cli/package.json
new file mode 100644
index 0000000..3a0397d
--- /dev/null
+++ b/cli/package.json
@@ -0,0 +1,37 @@
+{
+  "name": "cli",
+  "version": "1.0.0",
+  "description": "Command-line interface for Documind extractor",
+  "type": "module",
+  "main": "dist/index.js",
+  "types": "dist/index.d.ts",
+  "bin": {
+    "documind": "./dist/index.js"
+  },
+  "files": [
+    "dist"
+  ],
+  "scripts": {
+    "build": "tsc --build && tsc-alias -p tsconfig.json",
+    "start": "node dist/index.js",
+    "dev": "tsx src/index.ts"
+  },
+  "keywords": [
+    "cli",
+    "document",
+    "extraction",
+    "llm"
+  ],
+  "dependencies": {
+    "commander": "^12.0.0",
+    "chalk": "^5.3.0",
+    "ora": "^8.0.1",
+    "extractor": "*",
+    "dotenv": "^16.4.5"
+  },
+  "devDependencies": {
+    "@types/node": "^20.14.11",
+    "tsc-alias": "^1.8.8",
+    "typescript": "^5.6.3"
+  }
+}
diff --git a/cli/src/commands/convert.ts b/cli/src/commands/convert.ts
new file mode 100644
index 0000000..b277819
--- /dev/null
+++ b/cli/src/commands/convert.ts
@@ -0,0 +1,73 @@
+import { Command } from 'commander';
+import { formatter } from 'extractor';
+import ora from 'ora';
+import { logger } from '../utils/logger.js';
+import { fileExists, writeJsonFile, resolveFilePath } from '../utils/file-helper.js';
+import fs from 'fs';
+
+export function createConvertCommand(): Command {
+  const cmd = new Command('convert');
+
+  cmd
+    .description('Convert a document to markdown or plain text')
+    .requiredOption('-f, --file <path>', 'Path to the document file')
+    .option('-m, --model <model>', 'LLM model to use', 'gpt-4o-mini')
+    .option('-t, --format <format>', 'Output format (markdown or plaintext)', 'markdown')
+    .option('-o, --output <path>', 'Output file path')
+    .option('--base-url <url>', 'Base URL for local LLM (for Ollama)')
+    .action(async (options: ConvertOptions) => {
+      await handleConvert(options);
+    });
+
+  return cmd;
+}
+
+interface ConvertOptions {
+  file: string;
+  model: string;
+  format: 'markdown' | 'plaintext';
+  output?: string;
+  baseUrl?: string;
+}
+
+async function handleConvert(options: ConvertOptions): Promise<void> {
+  const spinner = ora('Starting conversion...').start();
+
+  try {
+    const filePath = resolveFilePath(options.file);
+    
+    if (!fileExists(filePath)) {
+      spinner.fail(`File not found: ${filePath}`);
+      process.exit(1);
+    }
+
+    if (options.baseUrl) {
+      process.env.BASE_URL = options.baseUrl;
+    }
+
+    spinner.text = `Converting document to ${options.format}...`;
+
+    let result: string;
+    
+    if (options.format === 'markdown') {
+      result = await formatter.markdown({ file: filePath, model: options.model });
+    } else {
+      result = await formatter.plaintext({ file: filePath, model: options.model });
+    }
+
+    spinner.succeed('Conversion completed!');
+
+    if (options.output) {
+      const outputPath = resolveFilePath(options.output);
+      fs.writeFileSync(outputPath, result, 'utf-8');
+      logger.success(`Output saved to: ${outputPath}`);
+    } else {
+      console.log('\n' + result);
+    }
+
+  } catch (error) {
+    spinner.fail('Conversion failed');
+    logger.error((error as Error).message);
+    process.exit(1);
+  }
+}
diff --git a/cli/src/commands/extract.ts b/cli/src/commands/extract.ts
new file mode 100644
index 0000000..20fb44b
--- /dev/null
+++ b/cli/src/commands/extract.ts
@@ -0,0 +1,147 @@
+import { Command } from 'commander';
+import { extract, type SchemaField, type ExtractResult } from 'extractor';
+import ora from 'ora';
+import { logger } from '../utils/logger.js';
+import { fileExists, readJsonFile, writeJsonFile, resolveFilePath } from '../utils/file-helper.js';
+import path from 'path';
+import chalk from 'chalk';
+
+export function createExtractCommand(): Command {
+  const cmd = new Command('extract');
+
+  cmd
+    .description('Extract structured data from a document')
+    .requiredOption('-f, --file <path>', 'Path to the document file')
+    .option('-m, --model <model>', 'LLM model to use (e.g., gpt-4o-mini, llama3.2-vision, gemini-1.5-flash)', 'gpt-4o-mini')
+    .option('-s, --schema <path>', 'Path to JSON schema file')
+    .option('-t, --template <name>', 'Name of a predefined template')
+    .option('-a, --auto-schema', 'Auto-generate schema from document')
+    .option('-i, --instructions <text>', 'Instructions for auto-schema generation')
+    .option('-o, --output <path>', 'Output file path for results (JSON)')
+    .option('--base-url <url>', 'Base URL for local LLM (for Ollama)')
+    .option('-l, --language <code>', 'Language code for OCR (e.g., eng, deu, fra)', 'eng')
+    .option('--image-quality <number>', 'Image quality for compression (1-100, lower=smaller file)', '85')
+    .option('--max-image-width <number>', 'Max width for image resizing in pixels', '2048')
+    .action(async (options) => {
+      await handleExtract(options);
+    });
+
+  return cmd;
+}
+
+interface ExtractOptions {
+  file: string;
+  model: string;
+  schema?: string;
+  template?: string;
+  autoSchema?: boolean;
+  instructions?: string;
+  output?: string;
+  baseUrl?: string;
+  language?: string;
+  imageQuality?: string;
+  maxImageWidth?: string;
+}
+
+async function handleExtract(options: ExtractOptions): Promise<void> {
+  const spinner = ora('Starting extraction...').start();
+
+  try {
+    // Resolve file path
+    const filePath = resolveFilePath(options.file);
+    
+    if (!fileExists(filePath)) {
+      spinner.fail(`File not found: ${filePath}`);
+      process.exit(1);
+    }
+
+    spinner.text = 'File found. Preparing extraction...';
+
+    // Set base URL for local LLM if provided
+    if (options.baseUrl) {
+      process.env.BASE_URL = options.baseUrl;
+      logger.info(`Using local LLM at: ${options.baseUrl}`);
+    }
+
+    // Prepare extraction parameters
+    let schema: SchemaField[] | undefined;
+    let autoSchemaOption: boolean | { instructions: string } | undefined;
+
+    if (options.schema) {
+      const schemaPath = resolveFilePath(options.schema);
+      if (!fileExists(schemaPath)) {
+        spinner.fail(`Schema file not found: ${schemaPath}`);
+        process.exit(1);
+      }
+      schema = readJsonFile(schemaPath);
+      spinner.text = 'Schema loaded from file';
+    }
+
+    if (options.autoSchema) {
+      if (options.instructions) {
+        autoSchemaOption = { instructions: options.instructions };
+        spinner.text = 'Auto-generating schema with instructions...';
+      } else {
+        autoSchemaOption = true;
+        spinner.text = 'Auto-generating schema...';
+      }
+    }
+
+    // Validate options
+    if (!schema && !options.template && !autoSchemaOption) {
+      spinner.fail('You must provide either --schema, --template, or --auto-schema');
+      process.exit(1);
+    }
+
+    spinner.text = `Extracting data using ${options.model}...`;
+
+    // Parse numeric options
+    const imageQuality = parseInt(options.imageQuality || '85', 10);
+    const maxImageWidth = parseInt(options.maxImageWidth || '2048', 10);
+
+    // Perform extraction
+    const result: ExtractResult = await extract({
+      file: filePath,
+      model: options.model,
+      schema,
+      template: options.template,
+      autoSchema: autoSchemaOption,
+      language: options.language || 'eng',
+      imageQuality,
+      maxImageWidth,
+    });
+
+    spinner.succeed('Extraction completed successfully!');
+
+    // Display results
+    logger.header('Extraction Results');
+    logger.info(`Document: ${result.fileName}`);
+    logger.info(`Pages processed: ${result.pages}`);
+    
+    console.log('\n' + chalk.bold('Extracted Data:'));
+    console.log(JSON.stringify(result.data, null, 2));
+
+    // Save to file if output path provided
+    if (options.output) {
+      const outputPath = resolveFilePath(options.output);
+      writeJsonFile(outputPath, {
+        fileName: result.fileName,
+        pages: result.pages,
+        data: result.data,
+        extractedAt: new Date().toISOString(),
+        model: options.model,
+      });
+      logger.success(`Results saved to: ${outputPath}`);
+    }
+
+  } catch (error) {
+    spinner.fail('Extraction failed');
+    logger.error((error as Error).message);
+    
+    if (process.env.DEBUG) {
+      console.error(error);
+    }
+    
+    process.exit(1);
+  }
+}
diff --git a/cli/src/commands/templates.ts b/cli/src/commands/templates.ts
new file mode 100644
index 0000000..9970bcd
--- /dev/null
+++ b/cli/src/commands/templates.ts
@@ -0,0 +1,39 @@
+import { Command } from 'commander';
+import { templates } from 'extractor';
+import { logger } from '../utils/logger.js';
+import chalk from 'chalk';
+
+export function createTemplatesCommand(): Command {
+  const cmd = new Command('templates');
+
+  cmd
+    .description('List available templates')
+    .action(() => {
+      handleTemplates();
+    });
+
+  return cmd;
+}
+
+function handleTemplates(): void {
+  try {
+    logger.header('Available Templates');
+    
+    const availableTemplates = templates.list();
+    
+    if (availableTemplates.length === 0) {
+      logger.warn('No templates found');
+      return;
+    }
+
+    availableTemplates.forEach((template: string) => {
+      console.log(chalk.cyan('  •'), chalk.bold(template));
+    });
+
+    console.log('\n' + chalk.dim('Use with: documind extract -f <file> -t <template-name>'));
+
+  } catch (error) {
+    logger.error(`Failed to list templates: ${(error as Error).message}`);
+    process.exit(1);
+  }
+}
diff --git a/cli/src/index.ts b/cli/src/index.ts
new file mode 100644
index 0000000..dae5642
--- /dev/null
+++ b/cli/src/index.ts
@@ -0,0 +1,66 @@
+#!/usr/bin/env node
+
+import { Command } from 'commander';
+import { config } from 'dotenv';
+import { createExtractCommand } from './commands/extract.js';
+import { createConvertCommand } from './commands/convert.js';
+import { createTemplatesCommand } from './commands/templates.js';
+import chalk from 'chalk';
+
+// Load environment variables
+config();
+
+const program = new Command();
+
+program
+  .name('documind')
+  .description('CLI tool for intelligent document processing and extraction')
+  .version('1.0.0');
+
+// Add commands
+program.addCommand(createExtractCommand());
+program.addCommand(createConvertCommand());
+program.addCommand(createTemplatesCommand());
+
+// Add help examples
+program.addHelpText('after', `
+${chalk.bold('Examples:')}
+
+  ${chalk.dim('# Extract data with auto-generated schema')}
+  ${chalk.cyan('$ documind extract -f invoice.pdf --auto-schema')}
+
+  ${chalk.dim('# Extract data with custom schema')}
+  ${chalk.cyan('$ documind extract -f document.pdf -s schema.json -o output.json')}
+
+  ${chalk.dim('# Extract data using a template')}
+  ${chalk.cyan('$ documind extract -f invoice.pdf -t invoice -o result.json')}
+
+  ${chalk.dim('# Extract data with local LLM (Ollama)')}
+  ${chalk.cyan('$ documind extract -f doc.pdf -m llama3.2-vision --base-url http://localhost:11434/v1 --auto-schema')}
+
+  ${chalk.dim('# Convert document to markdown')}
+  ${chalk.cyan('$ documind convert -f document.pdf -t markdown -o output.md')}
+
+  ${chalk.dim('# List available templates')}
+  ${chalk.cyan('$ documind templates')}
+
+${chalk.bold('Environment Variables:')}
+
+  ${chalk.yellow('OPENAI_API_KEY')}     - OpenAI API key for GPT models
+  ${chalk.yellow('GEMINI_API_KEY')}     - Google Gemini API key
+  ${chalk.yellow('BASE_URL')}           - Base URL for local LLM (Ollama)
+
+${chalk.bold('Supported Models:')}
+
+  ${chalk.green('OpenAI:')}   gpt-4o, gpt-4o-mini, gpt-4.1, gpt-4.1-mini
+  ${chalk.green('Google:')}   gemini-2.0-flash-001, gemini-1.5-flash, gemini-1.5-pro
+  ${chalk.green('Local:')}    llama3.2-vision (requires Ollama)
+`);
+
+// Parse arguments
+program.parse(process.argv);
+
+// Show help if no command provided
+if (!process.argv.slice(2).length) {
+  program.outputHelp();
+}
diff --git a/cli/src/utils/file-helper.ts b/cli/src/utils/file-helper.ts
new file mode 100644
index 0000000..e48473f
--- /dev/null
+++ b/cli/src/utils/file-helper.ts
@@ -0,0 +1,38 @@
+import fs from 'fs';
+import path from 'path';
+
+export function fileExists(filePath: string): boolean {
+  try {
+    return fs.existsSync(filePath);
+  } catch {
+    return false;
+  }
+}
+
+export function readJsonFile(filePath: string): any {
+  try {
+    const content = fs.readFileSync(filePath, 'utf-8');
+    return JSON.parse(content);
+  } catch (error) {
+    throw new Error(`Failed to read JSON file: ${(error as Error).message}`);
+  }
+}
+
+export function writeJsonFile(filePath: string, data: any): void {
+  try {
+    const dirPath = path.dirname(filePath);
+    if (!fs.existsSync(dirPath)) {
+      fs.mkdirSync(dirPath, { recursive: true });
+    }
+    fs.writeFileSync(filePath, JSON.stringify(data, null, 2), 'utf-8');
+  } catch (error) {
+    throw new Error(`Failed to write JSON file: ${(error as Error).message}`);
+  }
+}
+
+export function resolveFilePath(filePath: string): string {
+  if (path.isAbsolute(filePath)) {
+    return filePath;
+  }
+  return path.resolve(process.cwd(), filePath);
+}
diff --git a/cli/src/utils/logger.ts b/cli/src/utils/logger.ts
new file mode 100644
index 0000000..718c1b0
--- /dev/null
+++ b/cli/src/utils/logger.ts
@@ -0,0 +1,23 @@
+import chalk from 'chalk';
+
+export const logger = {
+  info: (message: string) => {
+    console.log(chalk.blue('ℹ'), message);
+  },
+  
+  success: (message: string) => {
+    console.log(chalk.green('✔'), message);
+  },
+  
+  error: (message: string) => {
+    console.error(chalk.red('✖'), message);
+  },
+  
+  warn: (message: string) => {
+    console.warn(chalk.yellow('⚠'), message);
+  },
+  
+  header: (message: string) => {
+    console.log(chalk.bold.cyan(`\n${message}\n${'='.repeat(message.length)}`));
+  }
+};
diff --git a/cli/tsconfig.json b/cli/tsconfig.json
new file mode 100644
index 0000000..3f3ca2f
--- /dev/null
+++ b/cli/tsconfig.json
@@ -0,0 +1,26 @@
+{
+  "compilerOptions": {
+    "target": "ES2020",
+    "module": "ES2020",
+    "moduleResolution": "node",
+    "declaration": true,
+    "declarationMap": true,
+    "outDir": "./dist",
+    "rootDir": "./src",
+    "strict": true,
+    "esModuleInterop": true,
+    "skipLibCheck": true,
+    "forceConsistentCasingInFileNames": true,
+    "resolveJsonModule": true,
+    "allowSyntheticDefaultImports": true,
+    "baseUrl": ".",
+    "paths": {
+      "extractor": ["../extractor/src/index.ts"]
+    }
+  },
+  "include": ["src/**/*"],
+  "exclude": ["node_modules", "dist"],
+  "references": [
+    { "path": "../extractor" }
+  ]
+}
diff --git a/core/dist/index.d.ts b/core/dist/index.d.ts
deleted file mode 100644
index 4db31a0..0000000
--- a/core/dist/index.d.ts
+++ /dev/null
@@ -1,2 +0,0 @@
-import { DocumindArgs, DocumindOutput } from "./types";
-export declare const documind: ({ cleanup, concurrency, filePath, llmParams, maintainFormat, model, outputDir, pagesToConvertAsImages, tempDir, }: DocumindArgs) => Promise<DocumindOutput>;
diff --git a/core/dist/index.js b/core/dist/index.js
deleted file mode 100644
index c0523f7..0000000
--- a/core/dist/index.js
+++ /dev/null
@@ -1,174 +0,0 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.documind = void 0;
-const utils_1 = require("./utils");
-const fs_extra_1 = __importDefault(require("fs-extra"));
-const os_1 = __importDefault(require("os"));
-const path_1 = __importDefault(require("path"));
-const p_limit_1 = __importDefault(require("p-limit"));
-const types_1 = require("./types");
-const providers_1 = require("./providers");
-const documind = async ({ cleanup = true, concurrency = 10, filePath, llmParams = {}, maintainFormat = false, model, //= ModelOptions.gpt_4o_mini,
-outputDir, pagesToConvertAsImages = -1, tempDir = os_1.default.tmpdir(), }) => {
-    let inputTokenCount = 0;
-    let outputTokenCount = 0;
-    let priorPage = "";
-    const aggregatedMarkdown = [];
-    const startTime = new Date();
-    // Basic checks
-    if (!filePath || !filePath.length) {
-        throw new Error("Missing file path");
-    }
-    const defaultModel = model ?? types_1.OpenAIModels.GPT_4O_MINI;
-    const validatedParams = (0, utils_1.validateLLMParams)(llmParams);
-    const providerInstance = providers_1.getModel.getProviderForModel(defaultModel);
-    // Ensure temp directory exists + create temp folder
-    const rand = Math.floor(1000 + Math.random() * 9000).toString();
-    const tempDirectory = path_1.default.join(tempDir || os_1.default.tmpdir(), `documind-file-${rand}`);
-    await fs_extra_1.default.ensureDir(tempDirectory);
-    // Download the PDF. Get file name.
-    const { extension, localPath } = await (0, utils_1.downloadFile)({
-        filePath,
-        tempDir: tempDirectory,
-    });
-    if (!localPath)
-        throw "Failed to save file to local drive";
-    // Sort the `pagesToConvertAsImages` array to make sure we use the right index
-    // for `formattedPages` as `pdf2pic` always returns images in order
-    if (Array.isArray(pagesToConvertAsImages)) {
-        pagesToConvertAsImages.sort((a, b) => a - b);
-    }
-    // Convert file to PDF if necessary
-    if (extension !== ".png") {
-        let pdfPath;
-        if (extension === ".pdf") {
-            pdfPath = localPath;
-        }
-        else {
-            pdfPath = await (0, utils_1.convertFileToPdf)({
-                extension,
-                localPath,
-                tempDir: tempDirectory,
-            });
-        }
-        // Convert the file to a series of images
-        await (0, utils_1.convertPdfToImages)({
-            localPath: pdfPath,
-            pagesToConvertAsImages,
-            tempDir: tempDirectory,
-        });
-    }
-    const endOfPath = localPath.split("/")[localPath.split("/").length - 1];
-    const rawFileName = endOfPath.split(".")[0];
-    const fileName = rawFileName
-        .replace(/[^\w\s]/g, "")
-        .replace(/\s+/g, "_")
-        .toLowerCase()
-        .substring(0, 255); // Truncate file name to 255 characters to prevent ENAMETOOLONG errors
-    // Get list of converted images
-    const files = await fs_extra_1.default.readdir(tempDirectory);
-    const images = files.filter((file) => file.endsWith(".png"));
-    if (maintainFormat) {
-        // Use synchronous processing
-        for (const image of images) {
-            const imagePath = path_1.default.join(tempDirectory, image);
-            try {
-                const { content, inputTokens, outputTokens } = await providerInstance.getCompletion({
-                    imagePath,
-                    llmParams: validatedParams,
-                    maintainFormat,
-                    model: defaultModel,
-                    priorPage,
-                });
-                const formattedMarkdown = (0, utils_1.formatMarkdown)(content);
-                inputTokenCount += inputTokens;
-                outputTokenCount += outputTokens;
-                // Update prior page to result from last processing step
-                priorPage = formattedMarkdown;
-                // Add all markdown results to array
-                aggregatedMarkdown.push(formattedMarkdown);
-            }
-            catch (error) {
-                console.error(`Failed to process image ${image}:`, error);
-                throw error;
-            }
-        }
-    }
-    else {
-        // Process in parallel with a limit on concurrent pages
-        const processPage = async (image) => {
-            const imagePath = path_1.default.join(tempDirectory, image);
-            try {
-                const { content, inputTokens, outputTokens } = await providerInstance.getCompletion({
-                    imagePath,
-                    llmParams: validatedParams,
-                    maintainFormat,
-                    model: defaultModel,
-                    priorPage,
-                });
-                const formattedMarkdown = (0, utils_1.formatMarkdown)(content);
-                inputTokenCount += inputTokens;
-                outputTokenCount += outputTokens;
-                // Update prior page to result from last processing step
-                priorPage = formattedMarkdown;
-                // Add all markdown results to array
-                return formattedMarkdown;
-            }
-            catch (error) {
-                console.error(`Failed to process image ${image}:`, error);
-                throw error;
-            }
-        };
-        // Function to process pages with concurrency limit
-        const processPagesInBatches = async (images, limit) => {
-            const results = [];
-            const promises = images.map((image, index) => limit(() => processPage(image).then((result) => {
-                results[index] = result;
-            })));
-            await Promise.all(promises);
-            return results;
-        };
-        const limit = (0, p_limit_1.default)(concurrency);
-        const results = await processPagesInBatches(images, limit);
-        const filteredResults = results.filter(utils_1.isString);
-        aggregatedMarkdown.push(...filteredResults);
-    }
-    // Write the aggregated markdown to a file
-    if (outputDir) {
-        const resultFilePath = path_1.default.join(outputDir, `${fileName}.md`);
-        await fs_extra_1.default.writeFile(resultFilePath, aggregatedMarkdown.join("\n\n"));
-    }
-    // Cleanup the downloaded PDF file
-    if (cleanup)
-        await fs_extra_1.default.remove(tempDirectory);
-    // Format JSON response
-    const endTime = new Date();
-    const completionTime = endTime.getTime() - startTime.getTime();
-    const formattedPages = aggregatedMarkdown.map((el, i) => {
-        let pageNumber;
-        // If we convert all pages, just use the array index
-        if (pagesToConvertAsImages === -1) {
-            pageNumber = i + 1;
-        }
-        // Else if we convert specific pages, use the page number from the parameter
-        else if (Array.isArray(pagesToConvertAsImages)) {
-            pageNumber = pagesToConvertAsImages[i];
-        }
-        // Else, the parameter is a number and use it for the page number
-        else {
-            pageNumber = pagesToConvertAsImages;
-        }
-        return { content: el, page: pageNumber, contentLength: el.length };
-    });
-    return {
-        completionTime,
-        fileName,
-        inputTokens: inputTokenCount,
-        outputTokens: outputTokenCount,
-        pages: formattedPages,
-    };
-};
-exports.documind = documind;
diff --git a/core/dist/openAI.d.ts b/core/dist/openAI.d.ts
deleted file mode 100644
index 2e390ed..0000000
--- a/core/dist/openAI.d.ts
+++ /dev/null
@@ -1,2 +0,0 @@
-import { CompletionArgs, CompletionResponse } from "./types";
-export declare const getCompletion: ({ apiKey, imagePath, llmParams, maintainFormat, model, priorPage, }: CompletionArgs) => Promise<CompletionResponse>;
diff --git a/core/dist/openAI.js b/core/dist/openAI.js
deleted file mode 100644
index 7c22118..0000000
--- a/core/dist/openAI.js
+++ /dev/null
@@ -1,75 +0,0 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.getCompletion = void 0;
-const utils_1 = require("./utils");
-const axios_1 = __importDefault(require("axios"));
-const getCompletion = async ({ apiKey, imagePath, llmParams, maintainFormat, model, priorPage, }) => {
-    const validModelsForCustomBaseUrl = [
-        "llava",
-        "llama3.2-vision",
-    ];
-    const validModelsForOpenAi = ["gpt-4o", "gpt-4o-mini"];
-    const baseUrl = process.env.BASE_URL || "https://api.openai.com/v1";
-    if (baseUrl !== "https://api.openai.com/v1") {
-        if (!validModelsForCustomBaseUrl.includes(model)) {
-            throw new Error(`Invalid model "${model}" for custom base URL. Valid options are: ${validModelsForCustomBaseUrl.join(", ")}.`);
-        }
-    }
-    else {
-        if (!validModelsForOpenAi.includes(model)) {
-            throw new Error(`Invalid model "${model}" for OpenAI. Valid options are: ${validModelsForOpenAi.join(", ")}.`);
-        }
-    }
-    const systemPrompt = `
-    Convert the following image/document  to markdown. 
-    Return only the markdown with no explanation text. Do not include deliminators like '''markdown.
-    You must include all information on the page. Do not exclude headers, footers, or subtext.
-  `;
-    // Default system message.
-    const messages = [{ role: "system", content: systemPrompt }];
-    // If content has already been generated, add it to context.
-    // This helps maintain the same format across pages
-    if (maintainFormat && priorPage && priorPage.length) {
-        messages.push({
-            role: "system",
-            content: `Markdown must maintain consistent formatting with the following page: \n\n """${priorPage}"""`,
-        });
-    }
-    // Add Image to request
-    const base64Image = await (0, utils_1.encodeImageToBase64)(imagePath);
-    messages.push({
-        role: "user",
-        content: [
-            {
-                type: "image_url",
-                image_url: { url: `data:image/png;base64,${base64Image}` },
-            },
-        ],
-    });
-    try {
-        const response = await axios_1.default.post(`${baseUrl}/chat/completions`, {
-            messages,
-            model,
-            ...(0, utils_1.convertKeysToSnakeCase)(llmParams ?? null),
-        }, {
-            headers: {
-                Authorization: `Bearer ${apiKey}`,
-                "Content-Type": "application/json",
-            },
-        });
-        const data = response.data;
-        return {
-            content: data.choices[0].message.content,
-            inputTokens: data.usage.prompt_tokens,
-            outputTokens: data.usage.completion_tokens,
-        };
-    }
-    catch (err) {
-        console.error("Error in OpenAI completion", err);
-        throw err;
-    }
-};
-exports.getCompletion = getCompletion;
diff --git a/core/dist/providers/google.d.ts b/core/dist/providers/google.d.ts
deleted file mode 100644
index da5acb0..0000000
--- a/core/dist/providers/google.d.ts
+++ /dev/null
@@ -1,5 +0,0 @@
-import { Completion } from "./utils/completion";
-import { CompletionArgs, CompletionResponse } from "../types";
-export declare class Google implements Completion {
-    getCompletion(args: CompletionArgs): Promise<CompletionResponse>;
-}
diff --git a/core/dist/providers/google.js b/core/dist/providers/google.js
deleted file mode 100644
index bf0e9d5..0000000
--- a/core/dist/providers/google.js
+++ /dev/null
@@ -1,71 +0,0 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.Google = void 0;
-const axios_1 = __importDefault(require("axios"));
-const types_1 = require("../types");
-const utils_1 = require("../utils");
-class Google {
-    async getCompletion(args) {
-        const { imagePath, llmParams, maintainFormat, model, priorPage, } = args;
-        if (!process.env.GEMINI_API_KEY) {
-            throw new Error("Missing GEMINI_API_KEY in environment variables.");
-        }
-        const apiKey = process.env.GEMINI_API_KEY;
-        const validModels = Object.values(types_1.GoogleModels);
-        if (!validModels.includes(model)) {
-            throw new Error(`Model "${model}" is not a google model.`);
-        }
-        const systemPrompt = `
-    Convert the following image/document  to markdown. 
-    Return only the markdown with no explanation text. Do not include deliminators like '''markdown.
-    You must include all information on the page. Do not exclude headers, footers, or subtext.
-  `;
-        const messages = [{ role: "system", content: systemPrompt }];
-        if (maintainFormat && priorPage) {
-            messages.push({
-                role: "system",
-                content: `Please ensure markdown formatting remains consistent with the prior page:\n\n"""${priorPage}"""`,
-            });
-        }
-        const base64Image = await (0, utils_1.encodeImageToBase64)(imagePath);
-        messages.push({
-            role: "user",
-            content: [
-                {
-                    "type": "text", //Using Gemini via openai requires text parameter
-                    "text": ""
-                },
-                {
-                    type: "image_url",
-                    image_url: { url: `data:image/png;base64,${base64Image}` },
-                },
-            ],
-        });
-        try {
-            const response = await axios_1.default.post("https://generativelanguage.googleapis.com/v1beta/openai/chat/completions", {
-                messages,
-                model,
-                // ...convertKeysToSnakeCase(llmParams ?? null),
-            }, {
-                headers: {
-                    Authorization: `Bearer ${apiKey}`,
-                    "Content-Type": "application/json",
-                },
-            });
-            const data = response.data;
-            return {
-                content: data.choices[0].message.content,
-                inputTokens: data.usage?.prompt_tokens ?? 0,
-                outputTokens: data.usage?.completion_tokens ?? 0,
-            };
-        }
-        catch (err) {
-            console.error("Google provider error:", err);
-            throw err;
-        }
-    }
-}
-exports.Google = Google;
diff --git a/core/dist/providers/index.d.ts b/core/dist/providers/index.d.ts
deleted file mode 100644
index 9503a8c..0000000
--- a/core/dist/providers/index.d.ts
+++ /dev/null
@@ -1,5 +0,0 @@
-import { Completion } from "./utils/completion";
-import { ModelOptions } from "../types";
-export declare class getModel {
-    static getProviderForModel(model: ModelOptions): Completion;
-}
diff --git a/core/dist/providers/index.js b/core/dist/providers/index.js
deleted file mode 100644
index de6439d..0000000
--- a/core/dist/providers/index.js
+++ /dev/null
@@ -1,22 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.getModel = void 0;
-const openAI_1 = require("./openAI");
-const ollama_1 = require("./ollama");
-const google_1 = require("./google");
-const types_1 = require("../types");
-class getModel {
-    static getProviderForModel(model) {
-        if (Object.values(types_1.OpenAIModels).includes(model)) {
-            return new openAI_1.OpenAI();
-        }
-        if (Object.values(types_1.GoogleModels).includes(model)) {
-            return new google_1.Google();
-        }
-        if (Object.values(types_1.LocalModels).includes(model)) {
-            return new ollama_1.Ollama();
-        }
-        throw new Error(`No provider found for model "${model}"`);
-    }
-}
-exports.getModel = getModel;
diff --git a/core/dist/providers/ollama.d.ts b/core/dist/providers/ollama.d.ts
deleted file mode 100644
index 3079696..0000000
--- a/core/dist/providers/ollama.d.ts
+++ /dev/null
@@ -1,5 +0,0 @@
-import { Completion } from "./utils/completion";
-import { CompletionArgs, CompletionResponse } from "../types";
-export declare class Ollama implements Completion {
-    getCompletion(args: CompletionArgs): Promise<CompletionResponse>;
-}
diff --git a/core/dist/providers/ollama.js b/core/dist/providers/ollama.js
deleted file mode 100644
index d80af13..0000000
--- a/core/dist/providers/ollama.js
+++ /dev/null
@@ -1,67 +0,0 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.Ollama = void 0;
-const axios_1 = __importDefault(require("axios"));
-const types_1 = require("../types");
-const utils_1 = require("../utils");
-class Ollama {
-    async getCompletion(args) {
-        const { imagePath, llmParams, maintainFormat, model, priorPage, } = args;
-        const baseUrl = process.env.BASE_URL;
-        if (!baseUrl) {
-            throw new Error("Missing BASE_URL in environment variables.");
-        }
-        const validModels = Object.values(types_1.LocalModels);
-        if (!validModels.includes(model)) {
-            throw new Error(`Model "${model}" is not a local model.`);
-        }
-        const systemPrompt = `
-    Convert the following image/document  to markdown. 
-    Return only the markdown with no explanation text. Do not include deliminators like '''markdown.
-    You must include all information on the page. Do not exclude headers, footers, or subtext.
-  `;
-        const messages = [{ role: "system", content: systemPrompt }];
-        if (maintainFormat && priorPage) {
-            messages.push({
-                role: "system",
-                content: `Please ensure markdown formatting remains consistent with the prior page:\n\n"""${priorPage}"""`,
-            });
-        }
-        const base64Image = await (0, utils_1.encodeImageToBase64)(imagePath);
-        messages.push({
-            role: "user",
-            content: [
-                {
-                    type: "image_url",
-                    image_url: { url: `data:image/png;base64,${base64Image}` },
-                },
-            ],
-        });
-        try {
-            const response = await axios_1.default.post(`${baseUrl}/chat/completions`, {
-                messages,
-                model,
-                ...(0, utils_1.convertKeysToSnakeCase)(llmParams ?? null),
-            }, {
-                headers: {
-                    // Authorization: "ollama",
-                    "Content-Type": "application/json",
-                },
-            });
-            const data = response.data;
-            return {
-                content: data.choices[0].message.content,
-                inputTokens: data.usage?.prompt_tokens ?? 0,
-                outputTokens: data.usage?.completion_tokens ?? 0,
-            };
-        }
-        catch (err) {
-            console.error("Local provider error:", err);
-            throw err;
-        }
-    }
-}
-exports.Ollama = Ollama;
diff --git a/core/dist/providers/openAI.d.ts b/core/dist/providers/openAI.d.ts
deleted file mode 100644
index 45e2f52..0000000
--- a/core/dist/providers/openAI.d.ts
+++ /dev/null
@@ -1,5 +0,0 @@
-import { Completion } from "./utils/completion";
-import { CompletionArgs, CompletionResponse } from "../types";
-export declare class OpenAI implements Completion {
-    getCompletion(args: CompletionArgs): Promise<CompletionResponse>;
-}
diff --git a/core/dist/providers/openAI.js b/core/dist/providers/openAI.js
deleted file mode 100644
index 8e1e649..0000000
--- a/core/dist/providers/openAI.js
+++ /dev/null
@@ -1,67 +0,0 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.OpenAI = void 0;
-const axios_1 = __importDefault(require("axios"));
-const types_1 = require("../types");
-const utils_1 = require("../utils");
-class OpenAI {
-    async getCompletion(args) {
-        const { imagePath, llmParams, maintainFormat, model, priorPage, } = args;
-        if (!process.env.OPENAI_API_KEY) {
-            throw new Error("Missing OPENAI_API_KEY in environment variables.");
-        }
-        const apiKey = process.env.OPENAI_API_KEY;
-        const validModels = Object.values(types_1.OpenAIModels);
-        if (!validModels.includes(model)) {
-            throw new Error(`Model "${model}" is not an OpenAI model.`);
-        }
-        const systemPrompt = `
-    Convert the following image/document  to markdown. 
-    Return only the markdown with no explanation text. Do not include deliminators like '''markdown.
-    You must include all information on the page. Do not exclude headers, footers, or subtext.
-  `;
-        const messages = [{ role: "system", content: systemPrompt }];
-        if (maintainFormat && priorPage) {
-            messages.push({
-                role: "system",
-                content: `Please ensure markdown formatting remains consistent with the prior page:\n\n"""${priorPage}"""`,
-            });
-        }
-        const base64Image = await (0, utils_1.encodeImageToBase64)(imagePath);
-        messages.push({
-            role: "user",
-            content: [
-                {
-                    type: "image_url",
-                    image_url: { url: `data:image/png;base64,${base64Image}` },
-                },
-            ],
-        });
-        try {
-            const response = await axios_1.default.post("https://api.openai.com/v1/chat/completions", {
-                messages,
-                model,
-                ...(0, utils_1.convertKeysToSnakeCase)(llmParams ?? null),
-            }, {
-                headers: {
-                    Authorization: `Bearer ${apiKey}`,
-                    "Content-Type": "application/json",
-                },
-            });
-            const data = response.data;
-            return {
-                content: data.choices[0].message.content,
-                inputTokens: data.usage?.prompt_tokens ?? 0,
-                outputTokens: data.usage?.completion_tokens ?? 0,
-            };
-        }
-        catch (err) {
-            console.error("OpenAI error:", err);
-            throw err;
-        }
-    }
-}
-exports.OpenAI = OpenAI;
diff --git a/core/dist/providers/utils/completion.d.ts b/core/dist/providers/utils/completion.d.ts
deleted file mode 100644
index 7753fb5..0000000
--- a/core/dist/providers/utils/completion.d.ts
+++ /dev/null
@@ -1,4 +0,0 @@
-import { CompletionArgs, CompletionResponse } from "../../types";
-export interface Completion {
-    getCompletion(args: CompletionArgs): Promise<CompletionResponse>;
-}
diff --git a/core/dist/providers/utils/completion.js b/core/dist/providers/utils/completion.js
deleted file mode 100644
index c8ad2e5..0000000
--- a/core/dist/providers/utils/completion.js
+++ /dev/null
@@ -1,2 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
diff --git a/core/dist/types.d.ts b/core/dist/types.d.ts
deleted file mode 100644
index 3f16a1c..0000000
--- a/core/dist/types.d.ts
+++ /dev/null
@@ -1,59 +0,0 @@
-export declare enum OpenAIModels {
-    GPT_4O = "gpt-4o",
-    GPT_4O_MINI = "gpt-4o-mini",
-    GPT_4_1 = "gpt-4.1",
-    GPT_4_1_MINI = "gpt-4.1-mini"
-}
-export declare enum LocalModels {
-    LLAMA3_2_VISION = "llama3.2-vision"
-}
-export declare enum GoogleModels {
-    GEMINI_2_FLASH = "gemini-2.0-flash-001",
-    GEMINI_2_FLASH_LITE = "gemini-2.0-flash-lite-preview-02-05",
-    GEMINI_1_5_FLASH = "gemini-1.5-flash",
-    GEMINI_1_5_FLASH_8B = "gemini-1.5-flash-8b",
-    GEMINI_1_5_PRO = "gemini-1.5-pro"
-}
-export type ModelOptions = OpenAIModels | GoogleModels | LocalModels;
-export interface DocumindArgs {
-    cleanup?: boolean;
-    concurrency?: number;
-    filePath: string;
-    llmParams?: LLMParams;
-    maintainFormat?: boolean;
-    model?: ModelOptions;
-    outputDir?: string;
-    pagesToConvertAsImages?: number | number[];
-    tempDir?: string;
-}
-export interface Page {
-    content: string;
-    contentLength: number;
-    page: number;
-}
-export interface DocumindOutput {
-    completionTime: number;
-    fileName: string;
-    inputTokens: number;
-    outputTokens: number;
-    pages: Page[];
-}
-export interface CompletionResponse {
-    content: string;
-    inputTokens: number;
-    outputTokens: number;
-}
-export interface CompletionArgs {
-    imagePath: string;
-    llmParams?: LLMParams;
-    maintainFormat: boolean;
-    model: ModelOptions;
-    priorPage: string;
-}
-export interface LLMParams {
-    frequencyPenalty?: number;
-    maxTokens?: number;
-    presencePenalty?: number;
-    temperature?: number;
-    topP?: number;
-}
diff --git a/core/dist/types.js b/core/dist/types.js
deleted file mode 100644
index 2828a4b..0000000
--- a/core/dist/types.js
+++ /dev/null
@@ -1,23 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.GoogleModels = exports.LocalModels = exports.OpenAIModels = void 0;
-var OpenAIModels;
-(function (OpenAIModels) {
-    OpenAIModels["GPT_4O"] = "gpt-4o";
-    OpenAIModels["GPT_4O_MINI"] = "gpt-4o-mini";
-    OpenAIModels["GPT_4_1"] = "gpt-4.1";
-    OpenAIModels["GPT_4_1_MINI"] = "gpt-4.1-mini";
-})(OpenAIModels || (exports.OpenAIModels = OpenAIModels = {}));
-var LocalModels;
-(function (LocalModels) {
-    //LLAVA = "llava",
-    LocalModels["LLAMA3_2_VISION"] = "llama3.2-vision";
-})(LocalModels || (exports.LocalModels = LocalModels = {}));
-var GoogleModels;
-(function (GoogleModels) {
-    GoogleModels["GEMINI_2_FLASH"] = "gemini-2.0-flash-001";
-    GoogleModels["GEMINI_2_FLASH_LITE"] = "gemini-2.0-flash-lite-preview-02-05";
-    GoogleModels["GEMINI_1_5_FLASH"] = "gemini-1.5-flash";
-    GoogleModels["GEMINI_1_5_FLASH_8B"] = "gemini-1.5-flash-8b";
-    GoogleModels["GEMINI_1_5_PRO"] = "gemini-1.5-pro";
-})(GoogleModels || (exports.GoogleModels = GoogleModels = {}));
diff --git a/core/dist/utils.d.ts b/core/dist/utils.d.ts
deleted file mode 100644
index aeab40c..0000000
--- a/core/dist/utils.d.ts
+++ /dev/null
@@ -1,27 +0,0 @@
-import { LLMParams } from "./types";
-export declare const validateLLMParams: (params: Partial<LLMParams>) => LLMParams;
-export declare const encodeImageToBase64: (imagePath: string) => Promise<string>;
-export declare const formatMarkdown: (text: string) => string;
-export declare const isString: (value: string | null) => value is string;
-export declare const isValidUrl: (string: string) => boolean;
-export declare const downloadFile: ({ filePath, tempDir, }: {
-    filePath: string;
-    tempDir: string;
-}) => Promise<{
-    extension: string;
-    localPath: string;
-}>;
-export declare const getTextFromImage: (buffer: Buffer) => Promise<{
-    confidence: number;
-}>;
-export declare const convertPdfToImages: ({ localPath, pagesToConvertAsImages, tempDir, }: {
-    localPath: string;
-    pagesToConvertAsImages: number | number[];
-    tempDir: string;
-}) => Promise<import("pdf2pic/dist/types/convertResponse").BufferResponse[]>;
-export declare const convertFileToPdf: ({ extension, localPath, tempDir, }: {
-    extension: string;
-    localPath: string;
-    tempDir: string;
-}) => Promise<string>;
-export declare const convertKeysToSnakeCase: (obj: Record<string, any> | null) => Record<string, any>;
diff --git a/core/dist/utils.js b/core/dist/utils.js
deleted file mode 100644
index 0885a39..0000000
--- a/core/dist/utils.js
+++ /dev/null
@@ -1,258 +0,0 @@
-"use strict";
-var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    var desc = Object.getOwnPropertyDescriptor(m, k);
-    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
-      desc = { enumerable: true, get: function() { return m[k]; } };
-    }
-    Object.defineProperty(o, k2, desc);
-}) : (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    o[k2] = m[k];
-}));
-var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
-    Object.defineProperty(o, "default", { enumerable: true, value: v });
-}) : function(o, v) {
-    o["default"] = v;
-});
-var __importStar = (this && this.__importStar) || function (mod) {
-    if (mod && mod.__esModule) return mod;
-    var result = {};
-    if (mod != null) for (var k in mod) if (k !== "default" && Object.prototype.hasOwnProperty.call(mod, k)) __createBinding(result, mod, k);
-    __setModuleDefault(result, mod);
-    return result;
-};
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.convertKeysToSnakeCase = exports.convertFileToPdf = exports.convertPdfToImages = exports.getTextFromImage = exports.downloadFile = exports.isValidUrl = exports.isString = exports.formatMarkdown = exports.encodeImageToBase64 = exports.validateLLMParams = void 0;
-const libreoffice_convert_1 = require("libreoffice-convert");
-const pdf2pic_1 = require("pdf2pic");
-const promises_1 = require("stream/promises");
-const util_1 = require("util");
-const Tesseract = __importStar(require("tesseract.js"));
-const axios_1 = __importDefault(require("axios"));
-const fs_extra_1 = __importDefault(require("fs-extra"));
-const mime_types_1 = __importDefault(require("mime-types"));
-const path_1 = __importDefault(require("path"));
-const sharp_1 = __importDefault(require("sharp"));
-const convertAsync = (0, util_1.promisify)(libreoffice_convert_1.convert);
-const defaultLLMParams = {
-    frequencyPenalty: 0, // OpenAI defaults to 0
-    maxTokens: 4000,
-    presencePenalty: 0, // OpenAI defaults to 0
-    temperature: 0,
-    topP: 1, // OpenAI defaults to 1
-};
-const validateLLMParams = (params) => {
-    const validKeys = Object.keys(defaultLLMParams);
-    for (const [key, value] of Object.entries(params)) {
-        if (!validKeys.includes(key)) {
-            throw new Error(`Invalid LLM parameter: ${key}`);
-        }
-        if (typeof value !== "number") {
-            throw new Error(`Value for '${key}' must be a number`);
-        }
-    }
-    return { ...defaultLLMParams, ...params };
-};
-exports.validateLLMParams = validateLLMParams;
-const encodeImageToBase64 = async (imagePath) => {
-    const imageBuffer = await fs_extra_1.default.readFile(imagePath);
-    return imageBuffer.toString("base64");
-};
-exports.encodeImageToBase64 = encodeImageToBase64;
-// Strip out the ```markdown wrapper
-const formatMarkdown = (text) => {
-    let formattedMarkdown = text?.trim();
-    let loopCount = 0;
-    const maxLoops = 3;
-    const startsWithMarkdown = formattedMarkdown.startsWith("```markdown");
-    while (startsWithMarkdown && loopCount < maxLoops) {
-        const endsWithClosing = formattedMarkdown.endsWith("```");
-        if (startsWithMarkdown && endsWithClosing) {
-            const outermostBlockRegex = /^```markdown\n([\s\S]*?)\n```$/;
-            const match = outermostBlockRegex.exec(formattedMarkdown);
-            if (match) {
-                formattedMarkdown = match[1].trim();
-                loopCount++;
-            }
-            else {
-                break;
-            }
-        }
-        else {
-            break;
-        }
-    }
-    return formattedMarkdown;
-};
-exports.formatMarkdown = formatMarkdown;
-const isString = (value) => {
-    return value !== null;
-};
-exports.isString = isString;
-const isValidUrl = (string) => {
-    let url;
-    try {
-        url = new URL(string);
-    }
-    catch (_) {
-        return false;
-    }
-    return url.protocol === "http:" || url.protocol === "https:";
-};
-exports.isValidUrl = isValidUrl;
-// Save file to local tmp directory
-const downloadFile = async ({ filePath, tempDir, }) => {
-    // Shorten the file name by removing URL parameters
-    const baseFileName = path_1.default.basename(filePath.split("?")[0]);
-    const localPath = path_1.default.join(tempDir, baseFileName);
-    let mimetype;
-    // Check if filePath is a URL
-    if ((0, exports.isValidUrl)(filePath)) {
-        const writer = fs_extra_1.default.createWriteStream(localPath);
-        const response = await (0, axios_1.default)({
-            url: filePath,
-            method: "GET",
-            responseType: "stream",
-        });
-        if (response.status !== 200) {
-            throw new Error(`HTTP error! Status: ${response.status}`);
-        }
-        mimetype = response.headers?.["content-type"];
-        await (0, promises_1.pipeline)(response.data, writer);
-    }
-    else {
-        // If filePath is a local file, copy it to the temp directory
-        await fs_extra_1.default.copyFile(filePath, localPath);
-    }
-    if (!mimetype) {
-        mimetype = mime_types_1.default.lookup(localPath);
-    }
-    let extension = mime_types_1.default.extension(mimetype) || "";
-    if (!extension) {
-        if (mimetype === "binary/octet-stream") {
-            extension = ".bin";
-        }
-        else {
-            throw new Error("File extension missing");
-        }
-    }
-    if (!extension.startsWith(".")) {
-        extension = `.${extension}`;
-    }
-    return { extension, localPath };
-};
-exports.downloadFile = downloadFile;
-// Extract text confidence from image buffer using Tesseract
-const getTextFromImage = async (buffer) => {
-    try {
-        // Get image and metadata
-        const image = (0, sharp_1.default)(buffer);
-        const metadata = await image.metadata();
-        // Crop to a 150px wide column in the center of the document.
-        // This section produced the highest confidence/speed tradeoffs.
-        const cropWidth = 150;
-        const cropHeight = metadata.height || 0;
-        const left = Math.max(0, Math.floor((metadata.width - cropWidth) / 2));
-        const top = 0;
-        // Extract the cropped image
-        const croppedBuffer = await image
-            .extract({ left, top, width: cropWidth, height: cropHeight })
-            .toBuffer();
-        // Pass the croppedBuffer to Tesseract.recognize
-        // @TODO: How can we generalize this to non eng languages?
-        const { data: { confidence }, } = await Tesseract.recognize(croppedBuffer, "eng");
-        return { confidence };
-    }
-    catch (error) {
-        console.error("Error during OCR:", error);
-        return { confidence: 0 };
-    }
-};
-exports.getTextFromImage = getTextFromImage;
-// Correct image orientation based on OCR confidence
-// Run Tesseract on 4 different orientations of the image and compare the output
-const correctImageOrientation = async (buffer) => {
-    const image = (0, sharp_1.default)(buffer);
-    const rotations = [0, 90, 180, 270];
-    const results = await Promise.all(rotations.map(async (rotation) => {
-        const rotatedImageBuffer = await image
-            .clone()
-            .rotate(rotation)
-            .toBuffer();
-        const { confidence } = await (0, exports.getTextFromImage)(rotatedImageBuffer);
-        return { rotation, confidence };
-    }));
-    // Find the rotation with the best confidence score
-    const bestResult = results.reduce((best, current) => current.confidence > best.confidence ? current : best);
-    if (bestResult.rotation !== 0) {
-        console.log(`Reorienting image ${bestResult.rotation} degrees (Confidence: ${bestResult.confidence}%).`);
-    }
-    // Rotate the image to the best orientation
-    const correctedImageBuffer = await image
-        .rotate(bestResult.rotation)
-        .toBuffer();
-    return correctedImageBuffer;
-};
-// Convert each page to a png, correct orientation, and save that image to tmp
-const convertPdfToImages = async ({ localPath, pagesToConvertAsImages, tempDir, }) => {
-    const options = {
-        density: 300,
-        format: "png",
-        height: 2048,
-        preserveAspectRatio: true,
-        saveFilename: path_1.default.basename(localPath, path_1.default.extname(localPath)),
-        savePath: tempDir,
-    };
-    const storeAsImage = (0, pdf2pic_1.fromPath)(localPath, options);
-    try {
-        const convertResults = await storeAsImage.bulk(pagesToConvertAsImages, {
-            responseType: "buffer",
-        });
-        await Promise.all(convertResults.map(async (result) => {
-            if (!result || !result.buffer) {
-                throw new Error("Could not convert page to image buffer");
-            }
-            if (!result.page)
-                throw new Error("Could not identify page data");
-            const paddedPageNumber = result.page.toString().padStart(5, "0");
-            // Correct the image orientation
-            const correctedBuffer = await correctImageOrientation(result.buffer);
-            const imagePath = path_1.default.join(tempDir, `${options.saveFilename}_page_${paddedPageNumber}.png`);
-            await fs_extra_1.default.writeFile(imagePath, correctedBuffer);
-        }));
-        return convertResults;
-    }
-    catch (err) {
-        console.error("Error during PDF conversion:", err);
-        throw err;
-    }
-};
-exports.convertPdfToImages = convertPdfToImages;
-// Convert each page (from other formats like docx) to a png and save that image to tmp
-const convertFileToPdf = async ({ extension, localPath, tempDir, }) => {
-    const inputBuffer = await fs_extra_1.default.readFile(localPath);
-    const outputFilename = path_1.default.basename(localPath, extension) + ".pdf";
-    const outputPath = path_1.default.join(tempDir, outputFilename);
-    try {
-        const pdfBuffer = await convertAsync(inputBuffer, ".pdf", undefined);
-        await fs_extra_1.default.writeFile(outputPath, pdfBuffer);
-        return outputPath;
-    }
-    catch (err) {
-        console.error(`Error converting ${extension} to .pdf:`, err);
-        throw err;
-    }
-};
-exports.convertFileToPdf = convertFileToPdf;
-const camelToSnakeCase = (str) => str.replace(/[A-Z]/g, (letter) => `_${letter.toLowerCase()}`);
-const convertKeysToSnakeCase = (obj) => {
-    if (typeof obj !== "object" || obj === null) {
-        return obj ?? {};
-    }
-    return Object.fromEntries(Object.entries(obj).map(([key, value]) => [camelToSnakeCase(key), value]));
-};
-exports.convertKeysToSnakeCase = convertKeysToSnakeCase;
diff --git a/core/package.json b/core/package.json
index 5efc90a..696b4e8 100644
--- a/core/package.json
+++ b/core/package.json
@@ -3,12 +3,13 @@
   "version": "1.0.0",
   "description": "Core package for document conversion",
   "main": "dist/index.js",
-  "type": "commonjs",
+  "types": "dist/index.d.ts",
+  "type": "module",
   "files": [
     "dist"
   ],
   "scripts": {
-    "build": "tsc"
+    "build": "tsc --build && tsc-alias -p tsconfig.json"
   },
   "dependencies": {
     "axios": "^1.7.2",
diff --git a/core/src/index.ts b/core/src/index.ts
index fd87076..001c399 100644
--- a/core/src/index.ts
+++ b/core/src/index.ts
@@ -5,7 +5,7 @@ import {
   formatMarkdown,
   isString,
   validateLLMParams
-} from "./utils";
+} from "./utils.js";
 import fs from "fs-extra";
 import os from "os";
 import path from "path";
@@ -15,9 +15,9 @@ import {
   DocumindOutput,
   ModelOptions,
   OpenAIModels,
-} from "./types";
-import { getModel } from "./providers";
-import { Completion } from "./providers/utils/completion";
+} from "./types.js";
+import { getModel } from "./providers/index.js";
+import { Completion } from "./providers/utils/completion.js";
 
 export const documind = async ({
   cleanup = true,
@@ -29,6 +29,9 @@ export const documind = async ({
   outputDir,
   pagesToConvertAsImages = -1,
   tempDir = os.tmpdir(),
+  language = "eng",
+  imageQuality = 85,
+  maxImageWidth = 2048,
 }: DocumindArgs): Promise<DocumindOutput> => {
 
   let inputTokenCount = 0;
@@ -83,6 +86,9 @@ export const documind = async ({
       localPath: pdfPath,
       pagesToConvertAsImages,
       tempDir: tempDirectory,
+      language,
+      imageQuality,
+      maxImageWidth,
     });
   }
 
@@ -94,9 +100,9 @@ export const documind = async ({
     .toLowerCase()
     .substring(0, 255); // Truncate file name to 255 characters to prevent ENAMETOOLONG errors
 
-  // Get list of converted images
+  // Get list of converted images (now in JPG format for better compression)
   const files = await fs.readdir(tempDirectory);
-  const images = files.filter((file) => file.endsWith(".png"));
+  const images = files.filter((file) => file.endsWith(".jpg") || file.endsWith(".png"));
 
   if (maintainFormat) {
     // Use synchronous processing
@@ -211,3 +217,6 @@ export const documind = async ({
     pages: formattedPages,
   };
 };
+
+export type { ModelOptions, DocumindArgs, DocumindOutput } from "./types.js";
+export { OpenAIModels, GoogleModels, LocalModels } from "./types.js";
diff --git a/core/src/providers/google.ts b/core/src/providers/google.ts
index 8052be8..febb15b 100644
--- a/core/src/providers/google.ts
+++ b/core/src/providers/google.ts
@@ -1,7 +1,7 @@
 import axios from "axios";
-import { Completion } from "./utils/completion";
-import { CompletionArgs, CompletionResponse, GoogleModels } from "../types";
-import { convertKeysToSnakeCase, encodeImageToBase64 } from "../utils";
+import { Completion } from "./utils/completion.js";
+import { CompletionArgs, CompletionResponse, GoogleModels } from "../types.js";
+import { convertKeysToSnakeCase, encodeImageToBase64 } from "../utils.js";
 
 export class Google implements Completion {
   public async getCompletion(args: CompletionArgs): Promise<CompletionResponse> {
diff --git a/core/src/providers/index.ts b/core/src/providers/index.ts
index 337d5fb..f3db33f 100644
--- a/core/src/providers/index.ts
+++ b/core/src/providers/index.ts
@@ -1,8 +1,8 @@
-import { Completion } from "./utils/completion";
-import { OpenAI } from "./openAI";
-import { Ollama } from "./ollama";
-import { Google } from "./google";
-import { ModelOptions, OpenAIModels, LocalModels, GoogleModels } from "../types";
+import { Completion } from "./utils/completion.js";
+import { OpenAI } from "./openAI.js";
+import { Ollama } from "./ollama.js";
+import { Google } from "./google.js";
+import { ModelOptions, OpenAIModels, LocalModels, GoogleModels } from "../types.js";
 
 export class getModel {
   public static getProviderForModel(model: ModelOptions): Completion {
diff --git a/core/src/providers/ollama.ts b/core/src/providers/ollama.ts
index ece3559..22ed612 100644
--- a/core/src/providers/ollama.ts
+++ b/core/src/providers/ollama.ts
@@ -1,7 +1,7 @@
 import axios from "axios";
-import { Completion } from "./utils/completion";
-import { CompletionArgs, CompletionResponse, LocalModels } from "../types";
-import { convertKeysToSnakeCase, encodeImageToBase64 } from "../utils";
+import { Completion } from "./utils/completion.js";
+import { CompletionArgs, CompletionResponse, LocalModels } from "../types.js";
+import { convertKeysToSnakeCase, encodeImageToBase64 } from "../utils.js";
 
 export class Ollama implements Completion {
   public async getCompletion(args: CompletionArgs): Promise<CompletionResponse> {
diff --git a/core/src/providers/openAI.ts b/core/src/providers/openAI.ts
index 90431fa..594a924 100644
--- a/core/src/providers/openAI.ts
+++ b/core/src/providers/openAI.ts
@@ -1,7 +1,7 @@
 import axios from "axios";
-import { Completion } from "./utils/completion";
-import { CompletionArgs, CompletionResponse, OpenAIModels } from "../types";
-import { convertKeysToSnakeCase, encodeImageToBase64 } from "../utils";
+import { Completion } from "./utils/completion.js";
+import { CompletionArgs, CompletionResponse, OpenAIModels } from "../types.js";
+import { convertKeysToSnakeCase, encodeImageToBase64 } from "../utils.js";
 
 export class OpenAI implements Completion {
   public async getCompletion(args: CompletionArgs): Promise<CompletionResponse> {
diff --git a/core/src/providers/utils/completion.ts b/core/src/providers/utils/completion.ts
index bbce5ba..3d2ac52 100644
--- a/core/src/providers/utils/completion.ts
+++ b/core/src/providers/utils/completion.ts
@@ -1,4 +1,4 @@
-import { CompletionArgs, CompletionResponse } from "../../types";
+import { CompletionArgs, CompletionResponse } from "../../types.js";
 
 export interface Completion {
   getCompletion(args: CompletionArgs): Promise<CompletionResponse>;
diff --git a/core/src/types.ts b/core/src/types.ts
index 67abc8b..25dbf59 100644
--- a/core/src/types.ts
+++ b/core/src/types.ts
@@ -6,8 +6,10 @@ export enum OpenAIModels {
 }
 
 export enum LocalModels {
-  //LLAVA = "llava",
+  LLAVA13 = "llava:13b",
+  LLAVA = "llava",
   LLAMA3_2_VISION = "llama3.2-vision",
+  QWEN2_5VL_7B = "qwen2.5vl:7b",
 }
 
 export enum GoogleModels {
@@ -30,6 +32,9 @@ export interface DocumindArgs {
   outputDir?: string;
   pagesToConvertAsImages?: number | number[];
   tempDir?: string;
+  language?: string; // Language code for OCR (e.g., 'eng', 'deu', 'fra')
+  imageQuality?: number; // Image quality for compression (1-100, default 85)
+  maxImageWidth?: number; // Max width for image resizing (default 2048)
 }
 
 export interface Page {
diff --git a/core/src/utils.ts b/core/src/utils.ts
index 9377fe4..e06e075 100644
--- a/core/src/utils.ts
+++ b/core/src/utils.ts
@@ -1,9 +1,9 @@
 import { convert } from "libreoffice-convert";
 import { fromPath } from "pdf2pic";
-import { LLMParams } from "./types";
+import { LLMParams } from "./types.js";
 import { pipeline } from "stream/promises";
 import { promisify } from "util";
-import * as Tesseract from "tesseract.js";
+import Tesseract from "tesseract.js";
 import axios from "axios";
 import fs from "fs-extra";
 import mime from "mime-types";
@@ -137,7 +137,8 @@ export const downloadFile = async ({
 
 // Extract text confidence from image buffer using Tesseract
 export const getTextFromImage = async (
-  buffer: Buffer
+  buffer: Buffer,
+  language: string = "eng"
 ): Promise<{ confidence: number }> => {
   try {
     // Get image and metadata
@@ -156,11 +157,10 @@ export const getTextFromImage = async (
       .extract({ left, top, width: cropWidth, height: cropHeight })
       .toBuffer();
 
-    // Pass the croppedBuffer to Tesseract.recognize
-    // @TODO: How can we generalize this to non eng languages?
+    // Pass the croppedBuffer to Tesseract.recognize with language parameter
     const {
       data: { confidence },
-    } = await Tesseract.recognize(croppedBuffer, "eng");
+    } = await Tesseract.recognize(croppedBuffer, language);
 
     return { confidence };
   } catch (error) {
@@ -171,7 +171,10 @@ export const getTextFromImage = async (
 
 // Correct image orientation based on OCR confidence
 // Run Tesseract on 4 different orientations of the image and compare the output
-const correctImageOrientation = async (buffer: Buffer): Promise<Buffer> => {
+const correctImageOrientation = async (
+  buffer: Buffer,
+  language: string = "eng"
+): Promise<Buffer> => {
   const image = sharp(buffer);
   const rotations = [0, 90, 180, 270];
 
@@ -181,7 +184,7 @@ const correctImageOrientation = async (buffer: Buffer): Promise<Buffer> => {
         .clone()
         .rotate(rotation)
         .toBuffer();
-      const { confidence } = await getTextFromImage(rotatedImageBuffer);
+      const { confidence } = await getTextFromImage(rotatedImageBuffer, language);
       return { rotation, confidence };
     })
   );
@@ -205,20 +208,27 @@ const correctImageOrientation = async (buffer: Buffer): Promise<Buffer> => {
   return correctedImageBuffer;
 };
 
-// Convert each page to a png, correct orientation, and save that image to tmp
+// Convert each page to a png, correct orientation, optimize, and save that image to tmp
 export const convertPdfToImages = async ({
   localPath,
   pagesToConvertAsImages,
   tempDir,
+  language = "eng",
+  imageQuality = 85,
+  maxImageWidth = 2048,
 }: {
   localPath: string;
   pagesToConvertAsImages: number | number[];
   tempDir: string;
+  language?: string;
+  imageQuality?: number;
+  maxImageWidth?: number;
 }) => {
+  // Use lower density (72 DPI is standard screen resolution, sufficient for LLM vision)
   const options = {
-    density: 300,
+    density: 72, // Reduced to 72 DPI to avoid high DPI warnings and reduce file size
     format: "png",
-    height: 2048,
+    width: maxImageWidth, // Use width instead of height for better control
     preserveAspectRatio: true,
     saveFilename: path.basename(localPath, path.extname(localPath)),
     savePath: tempDir,
@@ -237,14 +247,29 @@ export const convertPdfToImages = async ({
         if (!result.page) throw new Error("Could not identify page data");
         const paddedPageNumber = result.page.toString().padStart(5, "0");
 
-        // Correct the image orientation
-        const correctedBuffer = await correctImageOrientation(result.buffer);
-
+        // Correct the image orientation with language parameter
+        const correctedBuffer = await correctImageOrientation(result.buffer, language);
+
+        // Optimize image for LLM: resize and compress aggressively
+        // Convert to JPEG for better compression (much smaller than PNG)
+        const optimizedBuffer = await sharp(correctedBuffer)
+          .resize(maxImageWidth, null, {
+            fit: 'inside',
+            withoutEnlargement: true,
+          })
+          .jpeg({ 
+            quality: imageQuality, 
+            progressive: true,
+            mozjpeg: true // Use mozjpeg for better compression
+          })
+          .toBuffer();
+
+        // Save as JPG instead of PNG for better compression
         const imagePath = path.join(
           tempDir,
-          `${options.saveFilename}_page_${paddedPageNumber}.png`
+          `${options.saveFilename}_page_${paddedPageNumber}.jpg`
         );
-        await fs.writeFile(imagePath, correctedBuffer);
+        await fs.writeFile(imagePath, optimizedBuffer);
       })
     );
     return convertResults;
diff --git a/core/tsconfig.json b/core/tsconfig.json
index b962015..a96cc4b 100644
--- a/core/tsconfig.json
+++ b/core/tsconfig.json
@@ -1,12 +1,16 @@
 {
     "compilerOptions": {
       "target": "ES2020",
-      "module": "CommonJS",
+      "module": "ES2020",
+      "moduleResolution": "node",
       "declaration": true,
+      "declarationMap": true,
       "outDir": "./dist",
+      "rootDir": "./src",
       "strict": true,
       "esModuleInterop": true,
-      "skipLibCheck": true
+      "skipLibCheck": true,
+      "composite": true
     },
     "include": ["src/**/*"],
     "exclude": ["node_modules", "dist"]
diff --git a/extractor/package.json b/extractor/package.json
index 7953dd3..440346e 100644
--- a/extractor/package.json
+++ b/extractor/package.json
@@ -3,9 +3,15 @@
   "version": "1.0.0",
   "description": "Document extraction and processing package.",
   "type": "module",
+  "main": "dist/index.js",
+  "types": "dist/index.d.ts",
+  "files": [
+    "dist"
+  ],
   "scripts": {
-    "start": "node src/index.js",
-    "dev": "nodemon src/index.js"
+    "build": "tsc --build && tsc-alias -p tsconfig.json",
+    "start": "node dist/index.js",
+    "dev": "tsc --watch"
   },
   "dependencies": {
     "@ai-sdk/google": "^1.1.14",
@@ -21,6 +27,10 @@
     "zod-to-json-schema": "^3.24.2"
   },
   "devDependencies": {
-    "nodemon": "^3.1.7"
+    "@types/node": "^20.14.11",
+    "@types/uuid": "^11.0.0",
+    "nodemon": "^3.1.7",
+    "tsc-alias": "^1.8.8",
+    "typescript": "^5.6.3"
   }
 }
diff --git a/extractor/src/autoschema/autogenerateSchema.d.ts b/extractor/src/autoschema/autogenerateSchema.d.ts
new file mode 100644
index 0000000..b6d677c
--- /dev/null
+++ b/extractor/src/autoschema/autogenerateSchema.d.ts
@@ -0,0 +1,8 @@
+import { SchemaField } from '../utils/convertToZodSchema.js';
+type AutoSchemaOption = boolean | {
+    instructions: string;
+};
+export declare function autogenerateSchema(markdown: string, model: string, autoSchema: AutoSchemaOption): Promise<SchemaField[]>;
+export declare const autoschema: typeof autogenerateSchema;
+export {};
+//# sourceMappingURL=autogenerateSchema.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/autoschema/autogenerateSchema.d.ts.map b/extractor/src/autoschema/autogenerateSchema.d.ts.map
new file mode 100644
index 0000000..c6156d2
--- /dev/null
+++ b/extractor/src/autoschema/autogenerateSchema.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"autogenerateSchema.d.ts","sourceRoot":"","sources":["autogenerateSchema.ts"],"names":[],"mappings":"AAMA,OAAO,EAAE,WAAW,EAAE,MAAM,gCAAgC,CAAC;AAG7D,KAAK,gBAAgB,GAAG,OAAO,GAAG;IAAE,YAAY,EAAE,MAAM,CAAA;CAAE,CAAC;AAE3D,wBAAsB,kBAAkB,CACtC,QAAQ,EAAE,MAAM,EAChB,KAAK,EAAE,MAAM,EACb,UAAU,EAAE,gBAAgB,GAC3B,OAAO,CAAC,WAAW,EAAE,CAAC,CA0BxB;AAiED,eAAO,MAAM,UAAU,2BAAqB,CAAC"}
\ No newline at end of file
diff --git a/extractor/src/autoschema/autogenerateSchema.js b/extractor/src/autoschema/autogenerateSchema.js
index 46cfd0f..26bc9fc 100644
--- a/extractor/src/autoschema/autogenerateSchema.js
+++ b/extractor/src/autoschema/autogenerateSchema.js
@@ -5,93 +5,65 @@ import { baseSchema } from './generation-schemas/base.js';
 import { secondarySchema } from './generation-schemas/secondary.js';
 import { cleanSchemaFields } from "./cleanSchemaFields.js";
 import { z } from 'zod';
-
 export async function autogenerateSchema(markdown, model, autoSchema) {
-  if (autoSchema === true) {
-    return await blanketSchema(markdown, model);
-  }
-
-  if (
-    typeof autoSchema === "object" &&
-    autoSchema !== null
-  ) {
-    const keys = Object.keys(autoSchema);
-    if (keys.length !== 1 || keys[0] !== "instructions") {
-      throw new Error("autoSchema object must only have a single 'instructions' property");
+    if (autoSchema === true) {
+        return await blanketSchema(markdown, model);
     }
-
-    if (typeof autoSchema.instructions !== "string" || !autoSchema.instructions.trim()) {
-      throw new Error("Instructions can't be empty");
+    if (typeof autoSchema === "object" &&
+        autoSchema !== null) {
+        const keys = Object.keys(autoSchema);
+        if (keys.length !== 1 || keys[0] !== "instructions") {
+            throw new Error("autoSchema object must only have a single 'instructions' property");
+        }
+        if (typeof autoSchema.instructions !== "string" || !autoSchema.instructions.trim()) {
+            throw new Error("Instructions can't be empty");
+        }
+        return await instructionBasedSchema(markdown, model, autoSchema.instructions);
     }
-  
-
-    return await instructionBasedSchema(
-      markdown,
-      model,
-      autoSchema.instructions
-    );
-  }
-
-  return await blanketSchema(markdown, model);
+    return await blanketSchema(markdown, model);
 }
-
 async function blanketSchema(markdown, model) {
-  const extraction = getExtractor(model);
-  const schemaToUse = extraction === googleExtractor ? secondarySchema : baseSchema;
-  
-  const result = await extraction({
-    markdown,
-    zodSchema: schemaToUse,
-    prompt: AUTO_SCHEMA_PROMPT(markdown),
-    model: model, 
-  });
-
-  if (!result || !result.fields) {
-    throw new Error("Error auto generating default schema.");
-  }
-
-  return cleanSchemaFields(result.fields);
+    const extraction = getExtractor(model);
+    const schemaToUse = extraction === googleExtractor ? secondarySchema : baseSchema;
+    const result = await extraction({
+        markdown,
+        zodSchema: schemaToUse,
+        prompt: AUTO_SCHEMA_PROMPT(markdown),
+        model: model,
+    });
+    if (!result || !result.fields) {
+        throw new Error("Error auto generating default schema.");
+    }
+    return cleanSchemaFields(result.fields);
 }
-
 async function instructionBasedSchema(markdown, model, instructions) {
-
-  const instructionsZod = z.object({
-    fields: z.array(z.string()),
-  });
-
-  const instructionPrompt = `
+    const instructionsZod = z.object({
+        fields: z.array(z.string()),
+    });
+    const instructionPrompt = `
    Extract the name of the fields the user wants to extract.
-  `
-
-  const extraction = getExtractor(model);
-
-  const instructionFields = await extraction({
-    markdown: instructions,
-    zodSchema: instructionsZod,
-    prompt: instructionPrompt,
-    model: model,
-  });
-
-  if (!instructionFields || !instructionFields.fields) {
-    throw new Error("Error identifying the fields to be extracted.");
-  }
-
-  const data = instructionFields.fields;
-
-  const schemaToUse = extraction === googleExtractor ? secondarySchema : baseSchema;
-
-  const result = await extraction({
-    markdown,
-    zodSchema: schemaToUse,
-    prompt: INSTRUCTIONS_SCHEMA_PROMPT(markdown, data),
-    model: model,
-  });
-
-  if (!result || !result.fields) {
-    throw new Error("Error auto generating specified schema.");
-  }
-
-  return cleanSchemaFields(result.fields);
+  `;
+    const extraction = getExtractor(model);
+    const instructionFields = await extraction({
+        markdown: instructions,
+        zodSchema: instructionsZod,
+        prompt: instructionPrompt,
+        model: model,
+    });
+    if (!instructionFields || !instructionFields.fields) {
+        throw new Error("Error identifying the fields to be extracted.");
+    }
+    const data = instructionFields.fields;
+    const schemaToUse = extraction === googleExtractor ? secondarySchema : baseSchema;
+    const result = await extraction({
+        markdown,
+        zodSchema: schemaToUse,
+        prompt: INSTRUCTIONS_SCHEMA_PROMPT(markdown, data),
+        model: model,
+    });
+    if (!result || !result.fields) {
+        throw new Error("Error auto generating specified schema.");
+    }
+    return cleanSchemaFields(result.fields);
 }
-
-export const autoschema = autogenerateSchema;
\ No newline at end of file
+export const autoschema = autogenerateSchema;
diff --git a/extractor/src/autoschema/autogenerateSchema.ts b/extractor/src/autoschema/autogenerateSchema.ts
new file mode 100644
index 0000000..8d2844e
--- /dev/null
+++ b/extractor/src/autoschema/autogenerateSchema.ts
@@ -0,0 +1,107 @@
+import { getExtractor } from '../extractors/index.js';
+import { googleExtractor } from '../extractors/google.js';
+import { AUTO_SCHEMA_PROMPT, INSTRUCTIONS_SCHEMA_PROMPT } from "../prompts.js";
+import { baseSchema } from './generation-schemas/base.js';
+import { secondarySchema } from './generation-schemas/secondary.js';
+import { cleanSchemaFields } from "./cleanSchemaFields.js";
+import { SchemaField } from '../utils/convertToZodSchema.js';
+import { z } from 'zod';
+
+type AutoSchemaOption = boolean | { instructions: string };
+
+export async function autogenerateSchema(
+  markdown: string,
+  model: string,
+  autoSchema: AutoSchemaOption
+): Promise<SchemaField[]> {
+  if (autoSchema === true) {
+    return await blanketSchema(markdown, model);
+  }
+
+  if (
+    typeof autoSchema === "object" &&
+    autoSchema !== null
+  ) {
+    const keys = Object.keys(autoSchema);
+    if (keys.length !== 1 || keys[0] !== "instructions") {
+      throw new Error("autoSchema object must only have a single 'instructions' property");
+    }
+
+    if (typeof autoSchema.instructions !== "string" || !autoSchema.instructions.trim()) {
+      throw new Error("Instructions can't be empty");
+    }
+
+    return await instructionBasedSchema(
+      markdown,
+      model,
+      autoSchema.instructions
+    );
+  }
+
+  return await blanketSchema(markdown, model);
+}
+
+async function blanketSchema(markdown: string, model: string): Promise<SchemaField[]> {
+  const extraction = getExtractor(model);
+  const schemaToUse = extraction === googleExtractor ? secondarySchema : baseSchema;
+  
+  const result = await extraction({
+    markdown,
+    zodSchema: schemaToUse,
+    prompt: AUTO_SCHEMA_PROMPT(markdown),
+    model: model, 
+  });
+
+  if (!result || !result.fields) {
+    throw new Error("Error auto generating default schema.");
+  }
+
+  return cleanSchemaFields(result.fields);
+}
+
+async function instructionBasedSchema(
+  markdown: string,
+  model: string,
+  instructions: string
+): Promise<SchemaField[]> {
+
+  const instructionsZod = z.object({
+    fields: z.array(z.string()),
+  });
+
+  const instructionPrompt = `
+   Extract the name of the fields the user wants to extract.
+  `;
+
+  const extraction = getExtractor(model);
+
+  const instructionFields = await extraction({
+    markdown: instructions,
+    zodSchema: instructionsZod,
+    prompt: instructionPrompt,
+    model: model,
+  });
+
+  if (!instructionFields || !instructionFields.fields) {
+    throw new Error("Error identifying the fields to be extracted.");
+  }
+
+  const data = instructionFields.fields;
+
+  const schemaToUse = extraction === googleExtractor ? secondarySchema : baseSchema;
+
+  const result = await extraction({
+    markdown,
+    zodSchema: schemaToUse,
+    prompt: INSTRUCTIONS_SCHEMA_PROMPT(markdown, data),
+    model: model,
+  });
+
+  if (!result || !result.fields) {
+    throw new Error("Error auto generating specified schema.");
+  }
+
+  return cleanSchemaFields(result.fields);
+}
+
+export const autoschema = autogenerateSchema;
diff --git a/extractor/src/autoschema/cleanSchemaFields.d.ts b/extractor/src/autoschema/cleanSchemaFields.d.ts
new file mode 100644
index 0000000..2e23e52
--- /dev/null
+++ b/extractor/src/autoschema/cleanSchemaFields.d.ts
@@ -0,0 +1,3 @@
+import { SchemaField } from '../utils/convertToZodSchema.js';
+export declare function cleanSchemaFields(fields: SchemaField[]): SchemaField[];
+//# sourceMappingURL=cleanSchemaFields.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/autoschema/cleanSchemaFields.d.ts.map b/extractor/src/autoschema/cleanSchemaFields.d.ts.map
new file mode 100644
index 0000000..097aaca
--- /dev/null
+++ b/extractor/src/autoschema/cleanSchemaFields.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"cleanSchemaFields.d.ts","sourceRoot":"","sources":["cleanSchemaFields.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,WAAW,EAAE,MAAM,gCAAgC,CAAC;AAE7D,wBAAgB,iBAAiB,CAAC,MAAM,EAAE,WAAW,EAAE,GAAG,WAAW,EAAE,CASpE"}
\ No newline at end of file
diff --git a/extractor/src/autoschema/cleanSchemaFields.js b/extractor/src/autoschema/cleanSchemaFields.js
index 84e754c..3d0a7bc 100644
--- a/extractor/src/autoschema/cleanSchemaFields.js
+++ b/extractor/src/autoschema/cleanSchemaFields.js
@@ -1,10 +1,11 @@
 export function cleanSchemaFields(fields) {
     return fields.map((f) => {
-      if (f.children && f.children.length === 0) {
-        delete f.children;
-      } else if (f.children) {
-        f.children = cleanSchemaFields(f.children);
-      }
-      return f;
+        if (f.children && f.children.length === 0) {
+            delete f.children;
+        }
+        else if (f.children) {
+            f.children = cleanSchemaFields(f.children);
+        }
+        return f;
     });
-  }
\ No newline at end of file
+}
diff --git a/extractor/src/autoschema/cleanSchemaFields.ts b/extractor/src/autoschema/cleanSchemaFields.ts
new file mode 100644
index 0000000..a936a0d
--- /dev/null
+++ b/extractor/src/autoschema/cleanSchemaFields.ts
@@ -0,0 +1,12 @@
+import { SchemaField } from '../utils/convertToZodSchema.js';
+
+export function cleanSchemaFields(fields: SchemaField[]): SchemaField[] {
+    return fields.map((f) => {
+      if (f.children && f.children.length === 0) {
+        delete f.children;
+      } else if (f.children) {
+        f.children = cleanSchemaFields(f.children);
+      }
+      return f;
+    });
+  }
diff --git a/extractor/src/autoschema/generation-schemas/base.d.ts b/extractor/src/autoschema/generation-schemas/base.d.ts
new file mode 100644
index 0000000..0e1755b
--- /dev/null
+++ b/extractor/src/autoschema/generation-schemas/base.d.ts
@@ -0,0 +1,9 @@
+import { z } from "zod";
+export declare const baseSchema: z.ZodObject<{
+    fields: z.ZodArray<z.ZodType<any, z.ZodTypeDef, any>, "many">;
+}, "strip", z.ZodTypeAny, {
+    fields: any[];
+}, {
+    fields: any[];
+}>;
+//# sourceMappingURL=base.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/autoschema/generation-schemas/base.d.ts.map b/extractor/src/autoschema/generation-schemas/base.d.ts.map
new file mode 100644
index 0000000..8ce60da
--- /dev/null
+++ b/extractor/src/autoschema/generation-schemas/base.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"base.d.ts","sourceRoot":"","sources":["base.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAWxB,eAAO,MAAM,UAAU;;;;;;EAErB,CAAC"}
\ No newline at end of file
diff --git a/extractor/src/autoschema/generation-schemas/base.js b/extractor/src/autoschema/generation-schemas/base.js
index 80e144b..d8294f3 100644
--- a/extractor/src/autoschema/generation-schemas/base.js
+++ b/extractor/src/autoschema/generation-schemas/base.js
@@ -1,15 +1,11 @@
 // Used for non-google models that don't have schema limitations
 import { z } from "zod";
-
-const SchemaField = z.lazy(() =>
-  z.object({
+const SchemaField = z.lazy(() => z.object({
     name: z.string(),
     type: z.enum(["string", "number", "array", "object"]),
     description: z.string().optional(),
     children: z.array(SchemaField).optional(),
-  })
-);
-
+}));
 export const baseSchema = z.object({
-  fields: z.array(SchemaField),
-});
\ No newline at end of file
+    fields: z.array(SchemaField),
+});
diff --git a/extractor/src/autoschema/generation-schemas/base.ts b/extractor/src/autoschema/generation-schemas/base.ts
new file mode 100644
index 0000000..c367c60
--- /dev/null
+++ b/extractor/src/autoschema/generation-schemas/base.ts
@@ -0,0 +1,15 @@
+// Used for non-google models that don't have schema limitations
+import { z } from "zod";
+
+const SchemaField: z.ZodType<any> = z.lazy(() =>
+  z.object({
+    name: z.string(),
+    type: z.enum(["string", "number", "array", "object"]),
+    description: z.string().optional(),
+    children: z.array(SchemaField).optional(),
+  })
+);
+
+export const baseSchema = z.object({
+  fields: z.array(SchemaField),
+});
diff --git a/extractor/src/autoschema/generation-schemas/secondary.d.ts b/extractor/src/autoschema/generation-schemas/secondary.d.ts
new file mode 100644
index 0000000..c735a31
--- /dev/null
+++ b/extractor/src/autoschema/generation-schemas/secondary.d.ts
@@ -0,0 +1,63 @@
+import { z } from "zod";
+export declare const secondarySchema: z.ZodObject<{
+    fields: z.ZodArray<z.ZodObject<z.objectUtil.extendShape<{
+        name: z.ZodString;
+        type: z.ZodEnum<["string", "number", "array", "object"]>;
+        description: z.ZodOptional<z.ZodString>;
+    }, {
+        children: z.ZodOptional<z.ZodArray<z.ZodObject<{
+            name: z.ZodString;
+            type: z.ZodEnum<["string", "number", "array", "object"]>;
+            description: z.ZodOptional<z.ZodString>;
+        }, "strip", z.ZodTypeAny, {
+            type: "string" | "number" | "object" | "array";
+            name: string;
+            description?: string | undefined;
+        }, {
+            type: "string" | "number" | "object" | "array";
+            name: string;
+            description?: string | undefined;
+        }>, "many">>;
+    }>, "strip", z.ZodTypeAny, {
+        type: "string" | "number" | "object" | "array";
+        name: string;
+        description?: string | undefined;
+        children?: {
+            type: "string" | "number" | "object" | "array";
+            name: string;
+            description?: string | undefined;
+        }[] | undefined;
+    }, {
+        type: "string" | "number" | "object" | "array";
+        name: string;
+        description?: string | undefined;
+        children?: {
+            type: "string" | "number" | "object" | "array";
+            name: string;
+            description?: string | undefined;
+        }[] | undefined;
+    }>, "many">;
+}, "strip", z.ZodTypeAny, {
+    fields: {
+        type: "string" | "number" | "object" | "array";
+        name: string;
+        description?: string | undefined;
+        children?: {
+            type: "string" | "number" | "object" | "array";
+            name: string;
+            description?: string | undefined;
+        }[] | undefined;
+    }[];
+}, {
+    fields: {
+        type: "string" | "number" | "object" | "array";
+        name: string;
+        description?: string | undefined;
+        children?: {
+            type: "string" | "number" | "object" | "array";
+            name: string;
+            description?: string | undefined;
+        }[] | undefined;
+    }[];
+}>;
+//# sourceMappingURL=secondary.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/autoschema/generation-schemas/secondary.d.ts.map b/extractor/src/autoschema/generation-schemas/secondary.d.ts.map
new file mode 100644
index 0000000..a49cfdf
--- /dev/null
+++ b/extractor/src/autoschema/generation-schemas/secondary.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"secondary.d.ts","sourceRoot":"","sources":["secondary.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAYxB,eAAO,MAAM,eAAe;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;EAE1B,CAAC"}
\ No newline at end of file
diff --git a/extractor/src/autoschema/generation-schemas/secondary.js b/extractor/src/autoschema/generation-schemas/secondary.js
index 4a99c86..a8ea0eb 100644
--- a/extractor/src/autoschema/generation-schemas/secondary.js
+++ b/extractor/src/autoschema/generation-schemas/secondary.js
@@ -1,16 +1,13 @@
 // Used for google models
 import { z } from "zod";
-
 const BaseSchemaField = z.object({
-  name: z.string(),
-  type: z.enum(["string", "number", "array", "object"]),
-  description: z.string().optional(),
+    name: z.string(),
+    type: z.enum(["string", "number", "array", "object"]),
+    description: z.string().optional(),
 });
-
 const SchemaField = BaseSchemaField.extend({
-  children: z.array(BaseSchemaField).optional(),
+    children: z.array(BaseSchemaField).optional(),
 });
-
 export const secondarySchema = z.object({
-  fields: z.array(SchemaField),
-});
\ No newline at end of file
+    fields: z.array(SchemaField),
+});
diff --git a/extractor/src/autoschema/generation-schemas/secondary.ts b/extractor/src/autoschema/generation-schemas/secondary.ts
new file mode 100644
index 0000000..5d1baab
--- /dev/null
+++ b/extractor/src/autoschema/generation-schemas/secondary.ts
@@ -0,0 +1,16 @@
+// Used for google models
+import { z } from "zod";
+
+const BaseSchemaField = z.object({
+  name: z.string(),
+  type: z.enum(["string", "number", "array", "object"]),
+  description: z.string().optional(),
+});
+
+const SchemaField = BaseSchemaField.extend({
+  children: z.array(BaseSchemaField).optional(),
+});
+
+export const secondarySchema = z.object({
+  fields: z.array(SchemaField),
+});
diff --git a/extractor/src/converter.js b/extractor/src/converter.js
deleted file mode 100644
index 62df582..0000000
--- a/extractor/src/converter.js
+++ /dev/null
@@ -1,21 +0,0 @@
-import { documind } from 'core';
-import { generateMarkdownDocument } from './utils/generateMarkdown.js'; 
-
-export const convertFile = async (filePath, model) => {
-  try {
-    const result = await documind({
-      filePath,
-      model,
-    });
-
-    const { pages, fileName } = result;
-    const totalPages = pages.length;
-
-    const markdown = await generateMarkdownDocument(pages);
-    //console.log('Markdown generated', markdown);
-
-    return { markdown, totalPages, fileName };
-  } catch (error) {
-    console.error('Error running documind core:', error);
-  }
-};
diff --git a/extractor/src/converter.ts b/extractor/src/converter.ts
new file mode 100644
index 0000000..31929eb
--- /dev/null
+++ b/extractor/src/converter.ts
@@ -0,0 +1,37 @@
+import { documind, type ModelOptions } from 'core';
+import { generateMarkdownDocument } from './utils/generateMarkdown.js';
+
+interface ConvertFileResult {
+  markdown: string;
+  totalPages: number;
+  fileName: string;
+}
+
+export const convertFile = async (
+  filePath: string,
+  model: string,
+  language: string = "eng",
+  imageQuality: number = 85,
+  maxImageWidth: number = 2048
+): Promise<ConvertFileResult> => {
+  try {
+    const result = await documind({
+      filePath,
+      model: model as ModelOptions,
+      language,
+      imageQuality,
+      maxImageWidth,
+    });
+
+    const { pages, fileName } = result;
+    const totalPages = pages.length;
+
+    const markdown = await generateMarkdownDocument(pages);
+    //console.log('Markdown generated', markdown);
+
+    return { markdown, totalPages, fileName };
+  } catch (error) {
+    console.error('Error running documind core:', error);
+    throw error;
+  }
+};
diff --git a/extractor/src/extractors/google.d.ts b/extractor/src/extractors/google.d.ts
new file mode 100644
index 0000000..0103748
--- /dev/null
+++ b/extractor/src/extractors/google.d.ts
@@ -0,0 +1,10 @@
+import { ZodObject, ZodRawShape } from "zod";
+interface ExtractorParams {
+    markdown: string;
+    zodSchema: ZodObject<ZodRawShape>;
+    prompt: string;
+    model: string;
+}
+export declare const googleExtractor: ({ markdown, zodSchema, prompt, model }: ExtractorParams) => Promise<any>;
+export {};
+//# sourceMappingURL=google.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/extractors/google.d.ts.map b/extractor/src/extractors/google.d.ts.map
new file mode 100644
index 0000000..7959d61
--- /dev/null
+++ b/extractor/src/extractors/google.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"google.d.ts","sourceRoot":"","sources":["google.ts"],"names":[],"mappings":"AAEA,OAAO,EAAE,SAAS,EAAE,WAAW,EAAE,MAAM,KAAK,CAAC;AAE7C,UAAU,eAAe;IACvB,QAAQ,EAAE,MAAM,CAAC;IACjB,SAAS,EAAE,SAAS,CAAC,WAAW,CAAC,CAAC;IAClC,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;CACf;AAED,eAAO,MAAM,eAAe,2CAAkD,eAAe,KAAG,OAAO,CAAC,GAAG,CAyC1G,CAAA"}
\ No newline at end of file
diff --git a/extractor/src/extractors/google.js b/extractor/src/extractors/google.js
index 777fae1..7cb7e42 100644
--- a/extractor/src/extractors/google.js
+++ b/extractor/src/extractors/google.js
@@ -1,49 +1,36 @@
 import { GoogleGenerativeAI } from "@google/generative-ai";
 import { zodToJsonSchema } from "zod-to-json-schema";
-
-const genAI = new GoogleGenerativeAI(process.env.GEMINI_API_KEY);
-
 export const googleExtractor = async ({ markdown, zodSchema, prompt, model }) => {
     if (!process.env.GEMINI_API_KEY) {
-      throw new Error("Missing GEMINI_API_KEY");
+        throw new Error("Missing GEMINI_API_KEY");
     }
-
-const googleModel = model
-
-// Convert Zod schema to JSON schema
-let jsonSchema = zodToJsonSchema(zodSchema);
-
-// Remove additionalProperties and $schema keys
-const removeKeys = (obj) => {
-    if (Array.isArray(obj)) {
-        return obj.map(removeKeys);
-    } else if (typeof obj === "object" && obj !== null) {
-        return Object.fromEntries(
-            Object.entries(obj)
+    const genAI = new GoogleGenerativeAI(process.env.GEMINI_API_KEY);
+    const googleModel = model;
+    // Convert Zod schema to JSON schema
+    let jsonSchema = zodToJsonSchema(zodSchema);
+    // Remove additionalProperties and $schema keys
+    const removeKeys = (obj) => {
+        if (Array.isArray(obj)) {
+            return obj.map(removeKeys);
+        }
+        else if (typeof obj === "object" && obj !== null) {
+            return Object.fromEntries(Object.entries(obj)
                 .filter(([key]) => key !== "additionalProperties" && key !== "$schema")
-                .map(([key, value]) => [key, removeKeys(value)])
-        );
-    }
-    return obj;
-};
-
-jsonSchema = removeKeys(jsonSchema);
-
-const modelToUse = genAI.getGenerativeModel({
-    model: googleModel,
-    systemInstruction: prompt,
-    generationConfig: {
-      responseMimeType: "application/json",
-      responseSchema: jsonSchema,
-      },
+                .map(([key, value]) => [key, removeKeys(value)]));
+        }
+        return obj;
+    };
+    jsonSchema = removeKeys(jsonSchema);
+    const modelToUse = genAI.getGenerativeModel({
+        model: googleModel,
+        systemInstruction: prompt,
+        generationConfig: {
+            responseMimeType: "application/json",
+            responseSchema: jsonSchema,
+        },
     });
-    
-const result = await modelToUse.generateContent(
-    markdown,
-  );
-
-//console.log(result.response.text());
-const event = JSON.parse(result.response.text())
-return event;
-}
-
+    const result = await modelToUse.generateContent(markdown);
+    //console.log(result.response.text());
+    const event = JSON.parse(result.response.text());
+    return event;
+};
diff --git a/extractor/src/extractors/google.ts b/extractor/src/extractors/google.ts
new file mode 100644
index 0000000..c6648c1
--- /dev/null
+++ b/extractor/src/extractors/google.ts
@@ -0,0 +1,53 @@
+import { GoogleGenerativeAI } from "@google/generative-ai";
+import { zodToJsonSchema } from "zod-to-json-schema";
+import { ZodObject, ZodRawShape } from "zod";
+
+interface ExtractorParams {
+  markdown: string;
+  zodSchema: ZodObject<ZodRawShape>;
+  prompt: string;
+  model: string;
+}
+
+export const googleExtractor = async ({ markdown, zodSchema, prompt, model }: ExtractorParams): Promise<any> => {
+    if (!process.env.GEMINI_API_KEY) {
+      throw new Error("Missing GEMINI_API_KEY");
+    }
+
+    const genAI = new GoogleGenerativeAI(process.env.GEMINI_API_KEY);
+    const googleModel = model;
+
+    // Convert Zod schema to JSON schema
+    let jsonSchema = zodToJsonSchema(zodSchema);
+
+    // Remove additionalProperties and $schema keys
+    const removeKeys = (obj: any): any => {
+        if (Array.isArray(obj)) {
+            return obj.map(removeKeys);
+        } else if (typeof obj === "object" && obj !== null) {
+            return Object.fromEntries(
+                Object.entries(obj)
+                    .filter(([key]) => key !== "additionalProperties" && key !== "$schema")
+                    .map(([key, value]) => [key, removeKeys(value)])
+            );
+        }
+        return obj;
+    };
+
+    jsonSchema = removeKeys(jsonSchema);
+
+    const modelToUse = genAI.getGenerativeModel({
+        model: googleModel,
+        systemInstruction: prompt,
+        generationConfig: {
+          responseMimeType: "application/json",
+          responseSchema: jsonSchema as any,
+        },
+    });
+    
+    const result = await modelToUse.generateContent(markdown);
+
+    //console.log(result.response.text());
+    const event = JSON.parse(result.response.text());
+    return event;
+}
diff --git a/extractor/src/extractors/index.d.ts b/extractor/src/extractors/index.d.ts
new file mode 100644
index 0000000..c34a6a7
--- /dev/null
+++ b/extractor/src/extractors/index.d.ts
@@ -0,0 +1,13 @@
+import { ZodObject, ZodRawShape } from "zod";
+export declare const OpenAIModels: string[];
+export declare const LocalModels: string[];
+export declare const GoogleModels: string[];
+export interface ExtractorParams {
+    markdown: string;
+    zodSchema: ZodObject<ZodRawShape>;
+    prompt: string;
+    model: string;
+}
+export type ExtractorFunction = (params: ExtractorParams) => Promise<any>;
+export declare function getExtractor(model: string): ExtractorFunction;
+//# sourceMappingURL=index.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/extractors/index.d.ts.map b/extractor/src/extractors/index.d.ts.map
new file mode 100644
index 0000000..d03881d
--- /dev/null
+++ b/extractor/src/extractors/index.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["index.ts"],"names":[],"mappings":"AAGA,OAAO,EAAE,SAAS,EAAE,WAAW,EAAE,MAAM,KAAK,CAAC;AAE7C,eAAO,MAAM,YAAY,UAAuD,CAAC;AACjF,eAAO,MAAM,WAAW,UAAsB,CAAC;AAC/C,eAAO,MAAM,YAAY,UAMxB,CAAC;AAEF,MAAM,WAAW,eAAe;IAC9B,QAAQ,EAAE,MAAM,CAAC;IACjB,SAAS,EAAE,SAAS,CAAC,WAAW,CAAC,CAAC;IAClC,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;CACf;AAED,MAAM,MAAM,iBAAiB,GAAG,CAAC,MAAM,EAAE,eAAe,KAAK,OAAO,CAAC,GAAG,CAAC,CAAC;AAE1E,wBAAgB,YAAY,CAAC,KAAK,EAAE,MAAM,GAAG,iBAAiB,CAc7D"}
\ No newline at end of file
diff --git a/extractor/src/extractors/index.js b/extractor/src/extractors/index.js
index 485ad2a..b4bb0c2 100644
--- a/extractor/src/extractors/index.js
+++ b/extractor/src/extractors/index.js
@@ -1,29 +1,24 @@
 import { ollamaExtractor } from "./ollama.js";
 import { openAIExtractor } from "./openAI.js";
 import { googleExtractor } from "./google.js";
-
 export const OpenAIModels = ["gpt-4o", "gpt-4o-mini", "gpt-4.1", "gpt-4.1-mini"];
 export const LocalModels = ["llama3.2-vision"];
 export const GoogleModels = [
-  "gemini-2.0-flash-001", 
-  "gemini-2.0-flash-lite-preview-02-05", 
-  "gemini-1.5-flash",
-  "gemini-1.5-flash-8b",
- "gemini-1.5-pro"
+    "gemini-2.0-flash-001",
+    "gemini-2.0-flash-lite-preview-02-05",
+    "gemini-1.5-flash",
+    "gemini-1.5-flash-8b",
+    "gemini-1.5-pro"
 ];
-
 export function getExtractor(model) {
-  if (OpenAIModels.includes(model)) {
-    return openAIExtractor;
-  }
-
-  if (GoogleModels.includes(model)) {
-    return googleExtractor;
-  }
-
-  if (LocalModels.includes(model)) {
-    return ollamaExtractor;
-  }
-
-  throw new Error(`Unrecognised model '${model}'.`);
+    if (OpenAIModels.includes(model)) {
+        return openAIExtractor;
+    }
+    if (GoogleModels.includes(model)) {
+        return googleExtractor;
+    }
+    if (LocalModels.includes(model)) {
+        return ollamaExtractor;
+    }
+    throw new Error(`Unrecognised model '${model}'.`);
 }
diff --git a/extractor/src/extractors/index.ts b/extractor/src/extractors/index.ts
new file mode 100644
index 0000000..95411ba
--- /dev/null
+++ b/extractor/src/extractors/index.ts
@@ -0,0 +1,39 @@
+import { ollamaExtractor } from "./ollama.js";
+import { openAIExtractor } from "./openAI.js";
+import { googleExtractor } from "./google.js";
+import { ZodObject, ZodRawShape } from "zod";
+
+export const OpenAIModels = ["gpt-4o", "gpt-4o-mini", "gpt-4.1", "gpt-4.1-mini"];
+export const LocalModels = ["llama3.2-vision", "llava", "llava:13b", "qwen2.5vl:7b"];
+export const GoogleModels = [
+  "gemini-2.0-flash-001", 
+  "gemini-2.0-flash-lite-preview-02-05", 
+  "gemini-1.5-flash",
+  "gemini-1.5-flash-8b",
+  "gemini-1.5-pro"
+];
+
+export interface ExtractorParams {
+  markdown: string;
+  zodSchema: ZodObject<ZodRawShape>;
+  prompt: string;
+  model: string;
+}
+
+export type ExtractorFunction = (params: ExtractorParams) => Promise<any>;
+
+export function getExtractor(model: string): ExtractorFunction {
+  if (OpenAIModels.includes(model)) {
+    return openAIExtractor;
+  }
+
+  if (GoogleModels.includes(model)) {
+    return googleExtractor;
+  }
+
+  if (LocalModels.includes(model)) {
+    return ollamaExtractor;
+  }
+
+  throw new Error(`Unrecognised model '${model}'.`);
+}
diff --git a/extractor/src/extractors/ollama.d.ts b/extractor/src/extractors/ollama.d.ts
new file mode 100644
index 0000000..e1019a9
--- /dev/null
+++ b/extractor/src/extractors/ollama.d.ts
@@ -0,0 +1,10 @@
+import { ZodObject, ZodRawShape } from "zod";
+interface ExtractorParams {
+    markdown: string;
+    zodSchema: ZodObject<ZodRawShape>;
+    prompt: string;
+    model: string;
+}
+export declare const ollamaExtractor: ({ markdown, zodSchema, prompt, model }: ExtractorParams) => Promise<any>;
+export {};
+//# sourceMappingURL=ollama.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/extractors/ollama.d.ts.map b/extractor/src/extractors/ollama.d.ts.map
new file mode 100644
index 0000000..1a12e7b
--- /dev/null
+++ b/extractor/src/extractors/ollama.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"ollama.d.ts","sourceRoot":"","sources":["ollama.ts"],"names":[],"mappings":"AAEA,OAAO,EAAE,SAAS,EAAE,WAAW,EAAE,MAAM,KAAK,CAAC;AAE7C,UAAU,eAAe;IACvB,QAAQ,EAAE,MAAM,CAAC;IACjB,SAAS,EAAE,SAAS,CAAC,WAAW,CAAC,CAAC;IAClC,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;CACf;AAED,eAAO,MAAM,eAAe,2CAAkD,eAAe,KAAG,OAAO,CAAC,GAAG,CAuB1G,CAAA"}
\ No newline at end of file
diff --git a/extractor/src/extractors/ollama.js b/extractor/src/extractors/ollama.js
index a4fedc7..9868034 100644
--- a/extractor/src/extractors/ollama.js
+++ b/extractor/src/extractors/ollama.js
@@ -1,27 +1,22 @@
 import OpenAI from "openai";
 import { zodResponseFormat } from "openai/helpers/zod";
-
 export const ollamaExtractor = async ({ markdown, zodSchema, prompt, model }) => {
-  if (!process.env.BASE_URL) {
-    throw new Error("Missing BASE_URL");
-  }
-  
-  const openai = new OpenAI({ 
-    baseURL: process.env.BASE_URL,
-    apiKey: 'ollama'
-   });
-
-  const ollamaModel = model;
-
-  const completion = await openai.beta.chat.completions.parse({
-    model: ollamaModel,
-    messages: [
-      { role: "system", content: prompt },
-      { role: "user", content: markdown },
-    ],
-    response_format: zodResponseFormat(zodSchema, "event"),
-  });
-
-  const event = completion.choices[0].message.parsed;
-  return event;
-}
+    if (!process.env.BASE_URL) {
+        throw new Error("Missing BASE_URL");
+    }
+    const openai = new OpenAI({
+        baseURL: process.env.BASE_URL,
+        apiKey: 'ollama'
+    });
+    const ollamaModel = model;
+    const completion = await openai.beta.chat.completions.parse({
+        model: ollamaModel,
+        messages: [
+            { role: "system", content: prompt },
+            { role: "user", content: markdown },
+        ],
+        response_format: zodResponseFormat(zodSchema, "event"),
+    });
+    const event = completion.choices[0].message.parsed;
+    return event;
+};
diff --git a/extractor/src/extractors/ollama.ts b/extractor/src/extractors/ollama.ts
new file mode 100644
index 0000000..320bc20
--- /dev/null
+++ b/extractor/src/extractors/ollama.ts
@@ -0,0 +1,35 @@
+import OpenAI from "openai";
+import { zodResponseFormat } from "openai/helpers/zod";
+import { ZodObject, ZodRawShape } from "zod";
+
+interface ExtractorParams {
+  markdown: string;
+  zodSchema: ZodObject<ZodRawShape>;
+  prompt: string;
+  model: string;
+}
+
+export const ollamaExtractor = async ({ markdown, zodSchema, prompt, model }: ExtractorParams): Promise<any> => {
+  if (!process.env.BASE_URL) {
+    throw new Error("Missing BASE_URL");
+  }
+  
+  const openai = new OpenAI({ 
+    baseURL: process.env.BASE_URL,
+    apiKey: 'ollama'
+   });
+
+  const ollamaModel = model;
+
+  const completion = await openai.beta.chat.completions.parse({
+    model: ollamaModel,
+    messages: [
+      { role: "system", content: prompt },
+      { role: "user", content: markdown },
+    ],
+    response_format: zodResponseFormat(zodSchema, "event"),
+  });
+
+  const event = completion.choices[0].message.parsed;
+  return event;
+}
diff --git a/extractor/src/extractors/openAI.d.ts b/extractor/src/extractors/openAI.d.ts
new file mode 100644
index 0000000..7ec05d0
--- /dev/null
+++ b/extractor/src/extractors/openAI.d.ts
@@ -0,0 +1,10 @@
+import { ZodObject, ZodRawShape } from "zod";
+interface ExtractorParams {
+    markdown: string;
+    zodSchema: ZodObject<ZodRawShape>;
+    prompt: string;
+    model: string;
+}
+export declare const openAIExtractor: ({ markdown, zodSchema, prompt, model }: ExtractorParams) => Promise<any>;
+export {};
+//# sourceMappingURL=openAI.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/extractors/openAI.d.ts.map b/extractor/src/extractors/openAI.d.ts.map
new file mode 100644
index 0000000..ea66b3f
--- /dev/null
+++ b/extractor/src/extractors/openAI.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"openAI.d.ts","sourceRoot":"","sources":["openAI.ts"],"names":[],"mappings":"AAEA,OAAO,EAAK,SAAS,EAAE,WAAW,EAAE,MAAM,KAAK,CAAC;AAEhD,UAAU,eAAe;IACvB,QAAQ,EAAE,MAAM,CAAC;IACjB,SAAS,EAAE,SAAS,CAAC,WAAW,CAAC,CAAC;IAClC,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;CACf;AAED,eAAO,MAAM,eAAe,2CAAkD,eAAe,KAAG,OAAO,CAAC,GAAG,CAmB1G,CAAA"}
\ No newline at end of file
diff --git a/extractor/src/extractors/openAI.js b/extractor/src/extractors/openAI.js
index 62577db..ad33755 100644
--- a/extractor/src/extractors/openAI.js
+++ b/extractor/src/extractors/openAI.js
@@ -1,23 +1,19 @@
 import OpenAI from "openai";
 import { zodResponseFormat } from "openai/helpers/zod";
-
 export const openAIExtractor = async ({ markdown, zodSchema, prompt, model }) => {
-  if (!process.env.OPENAI_API_KEY) {
-    throw new Error("Missing OPENAI_API_KEY");
-  }
-
-  const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });
-  const openAiModel = model;
-
-  const completion = await openai.beta.chat.completions.parse({
-    model: openAiModel,
-    messages: [
-      { role: "system", content: prompt },
-      { role: "user", content: markdown },
-    ],
-    response_format: zodResponseFormat(zodSchema, "event"),
-  });
-
-  const event = completion.choices[0].message.parsed;
-  return event;
-}
+    if (!process.env.OPENAI_API_KEY) {
+        throw new Error("Missing OPENAI_API_KEY");
+    }
+    const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });
+    const openAiModel = model;
+    const completion = await openai.beta.chat.completions.parse({
+        model: openAiModel,
+        messages: [
+            { role: "system", content: prompt },
+            { role: "user", content: markdown },
+        ],
+        response_format: zodResponseFormat(zodSchema, "event"),
+    });
+    const event = completion.choices[0].message.parsed;
+    return event;
+};
diff --git a/extractor/src/extractors/openAI.ts b/extractor/src/extractors/openAI.ts
new file mode 100644
index 0000000..bac7ffb
--- /dev/null
+++ b/extractor/src/extractors/openAI.ts
@@ -0,0 +1,31 @@
+import OpenAI from "openai";
+import { zodResponseFormat } from "openai/helpers/zod";
+import { z, ZodObject, ZodRawShape } from "zod";
+
+interface ExtractorParams {
+  markdown: string;
+  zodSchema: ZodObject<ZodRawShape>;
+  prompt: string;
+  model: string;
+}
+
+export const openAIExtractor = async ({ markdown, zodSchema, prompt, model }: ExtractorParams): Promise<any> => {
+  if (!process.env.OPENAI_API_KEY) {
+    throw new Error("Missing OPENAI_API_KEY");
+  }
+
+  const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });
+  const openAiModel = model;
+
+  const completion = await openai.beta.chat.completions.parse({
+    model: openAiModel,
+    messages: [
+      { role: "system", content: prompt },
+      { role: "user", content: markdown },
+    ],
+    response_format: zodResponseFormat(zodSchema, "event"),
+  });
+
+  const event = completion.choices[0].message.parsed;
+  return event;
+}
diff --git a/extractor/src/index.js b/extractor/src/index.ts
similarity index 59%
rename from extractor/src/index.js
rename to extractor/src/index.ts
index 8c19a2d..8bbbdcd 100644
--- a/extractor/src/index.js
+++ b/extractor/src/index.ts
@@ -4,4 +4,7 @@ export { templates } from './services/templates.js';
 export { formatter } from './services/formatter.js';
 
 export { convertToZodSchema } from './utils/convertToZodSchema.js';
-export { getExtractor } from './extractors/index.js';
\ No newline at end of file
+export { getExtractor } from './extractors/index.js';
+
+export type { SchemaField } from './utils/convertToZodSchema.js';
+export type { ExtractOptions, ExtractResult } from './services/extract.js';
diff --git a/extractor/src/prompts.js b/extractor/src/prompts.ts
similarity index 89%
rename from extractor/src/prompts.js
rename to extractor/src/prompts.ts
index 7537a3f..2fd00f2 100644
--- a/extractor/src/prompts.js
+++ b/extractor/src/prompts.ts
@@ -6,16 +6,16 @@ You are an expert in structured data extraction. Your task is to extract informa
 - **Do not use substitutes such as "unknown," "missing," or any other placeholder for missing or unknown data. The value **must** always be explicitly null.
 `;
 
-export const AUTO_SCHEMA_PROMPT = (markdown) => `
+export const AUTO_SCHEMA_PROMPT = (markdown: string): string => `
 Read the following markdown content and generate a schema of useful structured data that can be extracted from it. Follow these rules strictly:
 - The \`children\` field **must only be present if the \`type\` is \`object\` or \`array\`. It should never exist for other types.
 - \`description\` fields should be concise, no longer than one sentence.
 """${markdown}"""
 `;
 
-export const INSTRUCTIONS_SCHEMA_PROMPT = (markdown, data) => `
+export const INSTRUCTIONS_SCHEMA_PROMPT = (markdown: string, data: string[]): string => `
 Read the following markdown content and generate a schema for the structured data I require: """${data}""". Use only the fields listed, and follow these rules strictly:
 - The \`children\` field **must only be present if the \`type\` is \`object\` or \`array\`. It should never exist for other types.
 - \`description\` fields should be concise, no longer than one sentence.
 """${markdown}"""
-`;
\ No newline at end of file
+`;
diff --git a/extractor/src/services/extract.d.ts b/extractor/src/services/extract.d.ts
new file mode 100644
index 0000000..f8d2095
--- /dev/null
+++ b/extractor/src/services/extract.d.ts
@@ -0,0 +1,24 @@
+import { SchemaField } from '../utils/convertToZodSchema.js';
+export interface ExtractOptions {
+    file: string;
+    schema?: SchemaField[];
+    template?: string;
+    model?: string;
+    autoSchema?: boolean | {
+        instructions?: string;
+    };
+}
+export interface ExtractResult {
+    success: boolean;
+    pages: number;
+    data: any;
+    fileName: string;
+    markdown: string;
+}
+/**
+ * Extracts data from a document based on a provided schema.
+ * @param options - Options for the extraction process.
+ * @returns - The result of the extraction, including pages, extracted data, and file name.
+ */
+export declare function extract({ file, schema, template, model, autoSchema }: ExtractOptions): Promise<ExtractResult>;
+//# sourceMappingURL=extract.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/services/extract.d.ts.map b/extractor/src/services/extract.d.ts.map
new file mode 100644
index 0000000..c33ae6f
--- /dev/null
+++ b/extractor/src/services/extract.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"extract.d.ts","sourceRoot":"","sources":["extract.ts"],"names":[],"mappings":"AAGA,OAAO,EAAsB,WAAW,EAAE,MAAM,gCAAgC,CAAC;AAMjF,MAAM,WAAW,cAAc;IAC7B,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,CAAC,EAAE,WAAW,EAAE,CAAC;IACvB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,KAAK,CAAC,EAAE,MAAM,CAAC;IACf,UAAU,CAAC,EAAE,OAAO,GAAG;QAAE,YAAY,CAAC,EAAE,MAAM,CAAA;KAAE,CAAC;CAClD;AAED,MAAM,WAAW,aAAa;IAC5B,OAAO,EAAE,OAAO,CAAC;IACjB,KAAK,EAAE,MAAM,CAAC;IACd,IAAI,EAAE,GAAG,CAAC;IACV,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC;CAClB;AAED;;;;GAIG;AACH,wBAAsB,OAAO,CAAC,EAAE,IAAI,EAAE,MAAM,EAAE,QAAQ,EAAE,KAAK,EAAE,UAAU,EAAE,EAAE,cAAc,GAAG,OAAO,CAAC,aAAa,CAAC,CA6DnH"}
\ No newline at end of file
diff --git a/extractor/src/services/extract.js b/extractor/src/services/extract.js
index add8284..1c3fee6 100644
--- a/extractor/src/services/extract.js
+++ b/extractor/src/services/extract.js
@@ -6,71 +6,63 @@ import { autogenerateSchema } from "../autoschema/autogenerateSchema.js";
 import { convertFile } from '../converter.js';
 import { BASE_EXTRACTION_PROMPT } from "../prompts.js";
 import { getExtractor } from '../extractors/index.js';
-
 /**
  * Extracts data from a document based on a provided schema.
- * @param {object} options - Options for the extraction process.
- * @param {string} options.file - The file path to the document.
- * @param {object} options.schema - The schema definition for data extraction.
- * @param {string} [options.template] - Name of a pre-defined template.
- * @param {string} [options.model] - The llm model to use if a base url is set.
- * @param {boolean | object} [options.autoSchema] - Option to auto-generate the schema.
- * @returns {Promise<object>} - The result of the extraction, including pages, extracted data, and file name.
+ * @param options - Options for the extraction process.
+ * @returns - The result of the extraction, including pages, extracted data, and file name.
  */
 export async function extract({ file, schema, template, model, autoSchema }) {
-  try {
-
-    const defaultModel = model || "gpt-4o-mini";
-
-    if (!file) {
-      throw new Error("File is required.");
+    try {
+        const defaultModel = model || "gpt-4o-mini";
+        if (!file) {
+            throw new Error("File is required.");
+        }
+        if (!isValidFile(file)) {
+            throw new Error("Invalid file type.");
+        }
+        let finalSchema = null;
+        if (template) {
+            finalSchema = getTemplate(template);
+        }
+        else if (schema) {
+            const { isValid, errors } = validateSchema(schema);
+            if (!isValid) {
+                throw new Error(`Invalid schema: ${errors.join(", ")}`);
+            }
+            finalSchema = schema;
+        }
+        else if (!autoSchema) {
+            throw new Error("You must provide a schema, template, or enable autoSchema.");
+        }
+        const { markdown, totalPages, fileName } = await convertFile(file, defaultModel);
+        if (autoSchema) {
+            const autoSchemaOption = autoSchema === true ? true : { instructions: autoSchema.instructions || "" };
+            finalSchema = await autogenerateSchema(markdown, defaultModel, autoSchemaOption);
+            if (!finalSchema) {
+                throw new Error("Failed to auto-generate schema.");
+            }
+        }
+        if (!finalSchema) {
+            throw new Error("No schema available for extraction.");
+        }
+        const dynamicZodSchema = convertToZodSchema(finalSchema);
+        const extraction = getExtractor(defaultModel);
+        const event = await extraction({
+            markdown,
+            zodSchema: dynamicZodSchema,
+            prompt: BASE_EXTRACTION_PROMPT,
+            model: defaultModel,
+        });
+        return {
+            success: true,
+            pages: totalPages,
+            data: event,
+            fileName,
+            markdown,
+        };
     }
-
-    if (!isValidFile(file)) {
-      throw new Error("Invalid file type.");
+    catch (error) {
+        console.error("Error processing document:", error);
+        throw new Error(`Failed to process document: ${error.message}`);
     }
-
-    let finalSchema = null;
-    if (template) {
-      finalSchema = getTemplate(template); 
-    } else if (schema) {
-      const { isValid, errors } = validateSchema(schema);
-      if (!isValid) {
-        throw new Error(`Invalid schema: ${errors.join(", ")}`);
-      }
-      finalSchema = schema;
-    } else if (!autoSchema) {
-      throw new Error("You must provide a schema, template, or enable autoSchema.");
-    }
-
-    const { markdown, totalPages, fileName } = await convertFile(file, defaultModel);
-
-    if (autoSchema) {
-      finalSchema = await autogenerateSchema(markdown, defaultModel, autoSchema); 
-      if (!finalSchema) {
-        throw new Error("Failed to auto-generate schema.");
-      }
-    }
-
-    const dynamicZodSchema = convertToZodSchema(finalSchema);
-    const extraction = getExtractor(defaultModel);
-
-    const event = await extraction({
-      markdown,
-      zodSchema: dynamicZodSchema,
-      prompt: BASE_EXTRACTION_PROMPT,
-      model: defaultModel,
-    });
-
-    return {
-      success: true,
-      pages: totalPages,
-      data: event,
-      fileName,
-      markdown,
-    };
-  } catch (error) {
-    console.error("Error processing document:", error);
-    throw new Error(`Failed to process document: ${error.message}`);
-  }
 }
diff --git a/extractor/src/services/extract.ts b/extractor/src/services/extract.ts
new file mode 100644
index 0000000..492c0a8
--- /dev/null
+++ b/extractor/src/services/extract.ts
@@ -0,0 +1,110 @@
+import { isValidFile } from '../utils/fileValidator.js';
+import { validateSchema } from '../utils/schemaValidator.js';
+import { getTemplate } from './templates.js';
+import { convertToZodSchema, SchemaField } from '../utils/convertToZodSchema.js';
+import { autogenerateSchema } from "../autoschema/autogenerateSchema.js";
+import { convertFile } from '../converter.js';
+import { BASE_EXTRACTION_PROMPT } from "../prompts.js";
+import { getExtractor } from '../extractors/index.js';
+
+export interface ExtractOptions {
+  file: string;
+  schema?: SchemaField[];
+  template?: string;
+  model?: string;
+  autoSchema?: boolean | { instructions?: string };
+  language?: string; // Language code for OCR (e.g., 'eng', 'deu', 'fra')
+  imageQuality?: number; // Image quality for compression (1-100, default 85)
+  maxImageWidth?: number; // Max width for image resizing (default 2048)
+}
+
+export interface ExtractResult {
+  success: boolean;
+  pages: number;
+  data: any;
+  fileName: string;
+  markdown: string;
+}
+
+/**
+ * Extracts data from a document based on a provided schema.
+ * @param options - Options for the extraction process.
+ * @returns - The result of the extraction, including pages, extracted data, and file name.
+ */
+export async function extract({ 
+  file, 
+  schema, 
+  template, 
+  model, 
+  autoSchema,
+  language = "eng",
+  imageQuality = 85,
+  maxImageWidth = 2048,
+}: ExtractOptions): Promise<ExtractResult> {
+  try {
+
+    const defaultModel = model || "gpt-4o-mini";
+
+    if (!file) {
+      throw new Error("File is required.");
+    }
+
+    if (!(await isValidFile(file))) {
+      throw new Error("Invalid file type.");
+    }
+
+    let finalSchema: SchemaField[] | null = null;
+    if (template) {
+      finalSchema = getTemplate(template); 
+    } else if (schema) {
+      const { isValid, errors } = validateSchema(schema);
+      if (!isValid) {
+        throw new Error(`Invalid schema: ${errors.join(", ")}`);
+      }
+      finalSchema = schema;
+    } else if (!autoSchema) {
+      throw new Error("You must provide a schema, template, or enable autoSchema.");
+    }
+
+    const { markdown, totalPages, fileName } = await convertFile(
+      file, 
+      defaultModel,
+      language,
+      imageQuality,
+      maxImageWidth
+    );
+
+    if (autoSchema) {
+      const autoSchemaOption = autoSchema === true ? true : { instructions: autoSchema.instructions || "" };
+      finalSchema = await autogenerateSchema(markdown, defaultModel, autoSchemaOption); 
+      if (!finalSchema) {
+        throw new Error("Failed to auto-generate schema.");
+      }
+    }
+
+    if (!finalSchema) {
+      throw new Error("No schema available for extraction.");
+    }
+
+    const dynamicZodSchema = convertToZodSchema(finalSchema);
+    const extraction = getExtractor(defaultModel);
+
+    const event = await extraction({
+      markdown,
+      zodSchema: dynamicZodSchema,
+      prompt: BASE_EXTRACTION_PROMPT,
+      model: defaultModel,
+    });
+
+    return {
+      success: true,
+      pages: totalPages,
+      data: event,
+      fileName,
+      markdown,
+    };
+  } catch (error) {
+    console.error("Error processing document:", error);
+    throw new Error(`Failed to process document: ${(error as Error).message}`);
+  }
+}
diff --git a/extractor/src/services/formatter.d.ts b/extractor/src/services/formatter.d.ts
new file mode 100644
index 0000000..e1071ad
--- /dev/null
+++ b/extractor/src/services/formatter.d.ts
@@ -0,0 +1,13 @@
+interface FormatterOptions {
+    file: string;
+    model?: string;
+}
+/**
+ * Formatter object for various formats.
+ */
+export declare const formatter: {
+    markdown: ({ file, model }: FormatterOptions) => Promise<string>;
+    plaintext: ({ file, model }: FormatterOptions) => Promise<string>;
+};
+export {};
+//# sourceMappingURL=formatter.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/services/formatter.d.ts.map b/extractor/src/services/formatter.d.ts.map
new file mode 100644
index 0000000..6cd09b4
--- /dev/null
+++ b/extractor/src/services/formatter.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"formatter.d.ts","sourceRoot":"","sources":["formatter.ts"],"names":[],"mappings":"AAIA,UAAU,gBAAgB;IACxB,IAAI,EAAE,MAAM,CAAC;IACb,KAAK,CAAC,EAAE,MAAM,CAAC;CAChB;AA8CD;;GAEG;AACH,eAAO,MAAM,SAAS;gCA1CsB,gBAAgB,KAAG,OAAO,CAAC,MAAM,CAAC;iCA6BjC,gBAAgB,KAAG,OAAO,CAAC,MAAM,CAAC;CAgB9E,CAAC"}
\ No newline at end of file
diff --git a/extractor/src/services/formatter.js b/extractor/src/services/formatter.js
index 07c395b..3c69578 100644
--- a/extractor/src/services/formatter.js
+++ b/extractor/src/services/formatter.js
@@ -1,58 +1,50 @@
 import { convertFile } from '../converter.js';
 import { isPdfFile } from '../utils/pdfValidator.js';
 import { convertToText } from '../utils/convertToText.js';
-
 /**
  * Extracts markdown content from a PDF.
- * @param {object} options - Options for extracting the markdown.
- * @param {string} options.file - The PDF file.
- * @param {string} [options.model] - The LLM model to use.
- * @returns {Promise<string>} - The markdown content.
+ * @param options - Options for extracting the markdown.
+ * @returns - The markdown content.
  */
 const getMarkdown = async ({ file, model }) => {
-  try {
-    if (!file) {
-        throw new Error('File is required.');
-      }
-  
-      if (!isPdfFile(file)) {
-        throw new Error('File must be a PDF.');
-      }
-
-    const { markdown } = await convertFile(file, model);
-
-    if (!markdown) {
-      throw new Error("Failed to extract markdown.");
+    try {
+        if (!file) {
+            throw new Error('File is required.');
+        }
+        const isPdf = await isPdfFile(file);
+        if (!isPdf) {
+            throw new Error('File must be a PDF.');
+        }
+        const { markdown } = await convertFile(file, model || "gpt-4o-mini");
+        if (!markdown) {
+            throw new Error("Failed to extract markdown.");
+        }
+        return markdown;
+    }
+    catch (error) {
+        console.error("Error extracting markdown:", error);
+        throw error;
     }
-
-    return markdown;
-  } catch (error) {
-    console.error("Error extracting markdown:", error);
-    throw error;
-  }
 };
-
 /**
  * Extracts plain text from a PDF by converting markdown to text.
- * @param {object} options - Options for extracting the plain text.
- * @param {string} options.file - The path to the PDF file.
- * @param {string} [options.model] - The LLM model to use.
- * @returns {Promise<string>} - The plain text content.
+ * @param options - Options for extracting the plain text.
+ * @returns - The plain text content.
  */
 const getPlainText = async ({ file, model }) => {
-  try {
-    const markdown = await getMarkdown({ file, model });
-    return convertToText(markdown);
-  } catch (error) {
-    console.error("Error extracting plain text:", error);
-    throw error;
-  }
+    try {
+        const markdown = await getMarkdown({ file, model });
+        return convertToText(markdown);
+    }
+    catch (error) {
+        console.error("Error extracting plain text:", error);
+        throw error;
+    }
 };
-
 /**
  * Formatter object for various formats.
  */
 export const formatter = {
-  markdown: getMarkdown,
-  plaintext: getPlainText,
+    markdown: getMarkdown,
+    plaintext: getPlainText,
 };
diff --git a/extractor/src/services/formatter.ts b/extractor/src/services/formatter.ts
new file mode 100644
index 0000000..d0fbca6
--- /dev/null
+++ b/extractor/src/services/formatter.ts
@@ -0,0 +1,60 @@
+import { convertFile } from '../converter.js';
+import { isPdfFile } from '../utils/pdfValidator.js';
+import { convertToText } from '../utils/convertToText.js';
+
+interface FormatterOptions {
+  file: string;
+  model?: string;
+}
+
+/**
+ * Extracts markdown content from a PDF.
+ * @param options - Options for extracting the markdown.
+ * @returns - The markdown content.
+ */
+const getMarkdown = async ({ file, model }: FormatterOptions): Promise<string> => {
+  try {
+    if (!file) {
+        throw new Error('File is required.');
+      }
+  
+      const isPdf = await isPdfFile(file);
+      if (!isPdf) {
+        throw new Error('File must be a PDF.');
+      }
+
+    const { markdown } = await convertFile(file, model || "gpt-4o-mini");
+
+    if (!markdown) {
+      throw new Error("Failed to extract markdown.");
+    }
+
+    return markdown;
+  } catch (error) {
+    console.error("Error extracting markdown:", error);
+    throw error;
+  }
+};
+
+/**
+ * Extracts plain text from a PDF by converting markdown to text.
+ * @param options - Options for extracting the plain text.
+ * @returns - The plain text content.
+ */
+const getPlainText = async ({ file, model }: FormatterOptions): Promise<string> => {
+  try {
+    const markdown = await getMarkdown({ file, model });
+    return convertToText(markdown);
+  } catch (error) {
+    console.error("Error extracting plain text:", error);
+    throw error;
+  }
+};
+
+/**
+ * Formatter object for various formats.
+ */
+export const formatter = {
+  markdown: getMarkdown,
+  plaintext: getPlainText,
+};
diff --git a/extractor/src/services/templates.d.ts b/extractor/src/services/templates.d.ts
new file mode 100644
index 0000000..8bb35b7
--- /dev/null
+++ b/extractor/src/services/templates.d.ts
@@ -0,0 +1,21 @@
+import { SchemaField } from '../utils/convertToZodSchema.js';
+/**
+ * Lists all available templates.
+ * @returns - Array of template names.
+ */
+export declare function listTemplates(): string[];
+/**
+ * Retrieves a specific template.
+ * @param name - The name of the template.
+ * @returns - The template content.
+ * @throws - If the template is not found.
+ */
+export declare function getTemplate(name: string): SchemaField[];
+/**
+ * Exports available templates.
+ */
+export declare const templates: {
+    list: typeof listTemplates;
+    get: typeof getTemplate;
+};
+//# sourceMappingURL=templates.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/services/templates.d.ts.map b/extractor/src/services/templates.d.ts.map
new file mode 100644
index 0000000..de1ddea
--- /dev/null
+++ b/extractor/src/services/templates.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"templates.d.ts","sourceRoot":"","sources":["templates.ts"],"names":[],"mappings":"AAGA,OAAO,EAAE,WAAW,EAAE,MAAM,gCAAgC,CAAC;AAO7D;;;GAGG;AACH,wBAAgB,aAAa,IAAI,MAAM,EAAE,CAKxC;AAED;;;;;GAKG;AACH,wBAAgB,WAAW,CAAC,IAAI,EAAE,MAAM,GAAG,WAAW,EAAE,CAMvD;AAED;;GAEG;AACH,eAAO,MAAM,SAAS;;;CAGrB,CAAC"}
\ No newline at end of file
diff --git a/extractor/src/services/templates.js b/extractor/src/services/templates.js
index dea34b5..1b6f29c 100644
--- a/extractor/src/services/templates.js
+++ b/extractor/src/services/templates.js
@@ -1,41 +1,36 @@
 import fs from 'fs';
 import * as path from 'path';
 import { fileURLToPath } from 'url';
-
 const __filename = fileURLToPath(import.meta.url);
 const __dirname = path.dirname(__filename);
-
 const templatesDirectory = path.resolve(__dirname, '../templates');
-
 /**
  * Lists all available templates.
- * @returns {string[]} - Array of template names.
+ * @returns - Array of template names.
  */
 export function listTemplates() {
-  return fs
-    .readdirSync(templatesDirectory)
-    .filter((file) => file.endsWith('.json'))
-    .map((file) => file.replace('.json', ''));
+    return fs
+        .readdirSync(templatesDirectory)
+        .filter((file) => file.endsWith('.json'))
+        .map((file) => file.replace('.json', ''));
 }
-
 /**
  * Retrieves a specific template.
- * @param {string} name - The name of the template.
- * @returns {object} - The template content.
- * @throws {Error} - If the template is not found.
+ * @param name - The name of the template.
+ * @returns - The template content.
+ * @throws - If the template is not found.
  */
 export function getTemplate(name) {
-  const templatePath = path.join(templatesDirectory, `${name}.json`);
-  if (!fs.existsSync(templatePath)) {
-    throw new Error(`Template "${name}" not found`);
-  }
-  return JSON.parse(fs.readFileSync(templatePath, 'utf8'));
+    const templatePath = path.join(templatesDirectory, `${name}.json`);
+    if (!fs.existsSync(templatePath)) {
+        throw new Error(`Template "${name}" not found`);
+    }
+    return JSON.parse(fs.readFileSync(templatePath, 'utf8'));
 }
-
 /**
  * Exports available templates.
  */
 export const templates = {
-  list: listTemplates,
-  get: getTemplate,
+    list: listTemplates,
+    get: getTemplate,
 };
diff --git a/extractor/src/services/templates.ts b/extractor/src/services/templates.ts
new file mode 100644
index 0000000..49c5f40
--- /dev/null
+++ b/extractor/src/services/templates.ts
@@ -0,0 +1,42 @@
+import fs from 'fs';
+import * as path from 'path';
+import { fileURLToPath } from 'url';
+import { SchemaField } from '../utils/convertToZodSchema.js';
+
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = path.dirname(__filename);
+
+const templatesDirectory = path.resolve(__dirname, '../templates');
+
+/**
+ * Lists all available templates.
+ * @returns - Array of template names.
+ */
+export function listTemplates(): string[] {
+  return fs
+    .readdirSync(templatesDirectory)
+    .filter((file) => file.endsWith('.json'))
+    .map((file) => file.replace('.json', ''));
+}
+
+/**
+ * Retrieves a specific template.
+ * @param name - The name of the template.
+ * @returns - The template content.
+ * @throws - If the template is not found.
+ */
+export function getTemplate(name: string): SchemaField[] {
+  const templatePath = path.join(templatesDirectory, `${name}.json`);
+  if (!fs.existsSync(templatePath)) {
+    throw new Error(`Template "${name}" not found`);
+  }
+  return JSON.parse(fs.readFileSync(templatePath, 'utf8'));
+}
+
+/**
+ * Exports available templates.
+ */
+export const templates = {
+  list: listTemplates,
+  get: getTemplate,
+};
diff --git a/extractor/src/utils/convertToText.d.ts b/extractor/src/utils/convertToText.d.ts
new file mode 100644
index 0000000..9623778
--- /dev/null
+++ b/extractor/src/utils/convertToText.d.ts
@@ -0,0 +1,2 @@
+export declare const convertToText: (markdown: string) => string;
+//# sourceMappingURL=convertToText.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/utils/convertToText.d.ts.map b/extractor/src/utils/convertToText.d.ts.map
new file mode 100644
index 0000000..f216108
--- /dev/null
+++ b/extractor/src/utils/convertToText.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"convertToText.d.ts","sourceRoot":"","sources":["convertToText.ts"],"names":[],"mappings":"AAAA,eAAO,MAAM,aAAa,aAAc,MAAM,KAAG,MAoB9C,CAAC"}
\ No newline at end of file
diff --git a/extractor/src/utils/convertToText.js b/extractor/src/utils/convertToText.js
index 8366f31..65c1219 100644
--- a/extractor/src/utils/convertToText.js
+++ b/extractor/src/utils/convertToText.js
@@ -1,22 +1,19 @@
 export const convertToText = (markdown) => {
     if (!markdown || typeof markdown !== "string") {
-      throw new Error("Valid markdown content is required.");
+        throw new Error("Valid markdown content is required.");
     }
-  
     // Strip markdown syntax and handle tables
     const plainText = markdown
-      .replace(/(\*\*|__)(.*?)\1/g, "$2") // Bold
-      .replace(/(\*|_)(.*?)\1/g, "$2")   // Italic
-      .replace(/(#+\s)/g, "")            // Headings
-      .replace(/\[(.*?)\]\(.*?\)/g, "$1") // Links
-      .replace(/!\[(.*?)\]\(.*?\)/g, "$1") // Images
-      .replace(/(```.*?\n[\s\S]*?\n```|`.*?`)/g, "") // Code blocks/inline
-      .replace(/>+/g, "")                // Blockquotes
-      .replace(/\n{2,}/g, "\n")          // Excess newlines
-      .replace(/\|([^|]*)\|/g, (_, row) => row.trim()) // Table rows
-      .replace(/-+/g, "")                // Table dividers (---|---)
-      .trim();
-  
+        .replace(/(\*\*|__)(.*?)\1/g, "$2") // Bold
+        .replace(/(\*|_)(.*?)\1/g, "$2") // Italic
+        .replace(/(#+\s)/g, "") // Headings
+        .replace(/\[(.*?)\]\(.*?\)/g, "$1") // Links
+        .replace(/!\[(.*?)\]\(.*?\)/g, "$1") // Images
+        .replace(/(```.*?\n[\s\S]*?\n```|`.*?`)/g, "") // Code blocks/inline
+        .replace(/>+/g, "") // Blockquotes
+        .replace(/\n{2,}/g, "\n") // Excess newlines
+        .replace(/\|([^|]*)\|/g, (_, row) => row.trim()) // Table rows
+        .replace(/-+/g, "") // Table dividers (---|---)
+        .trim();
     return plainText;
-  };
-  
\ No newline at end of file
+};
diff --git a/extractor/src/utils/convertToText.ts b/extractor/src/utils/convertToText.ts
new file mode 100644
index 0000000..7b960e0
--- /dev/null
+++ b/extractor/src/utils/convertToText.ts
@@ -0,0 +1,21 @@
+export const convertToText = (markdown: string): string => {
+    if (!markdown || typeof markdown !== "string") {
+      throw new Error("Valid markdown content is required.");
+    }
+  
+    // Strip markdown syntax and handle tables
+    const plainText = markdown
+      .replace(/(\*\*|__)(.*?)\1/g, "$2") // Bold
+      .replace(/(\*|_)(.*?)\1/g, "$2")   // Italic
+      .replace(/(#+\s)/g, "")            // Headings
+      .replace(/\[(.*?)\]\(.*?\)/g, "$1") // Links
+      .replace(/!\[(.*?)\]\(.*?\)/g, "$1") // Images
+      .replace(/(```.*?\n[\s\S]*?\n```|`.*?`)/g, "") // Code blocks/inline
+      .replace(/>+/g, "")                // Blockquotes
+      .replace(/\n{2,}/g, "\n")          // Excess newlines
+      .replace(/\|([^|]*)\|/g, (_, row) => row.trim()) // Table rows
+      .replace(/-+/g, "")                // Table dividers (---|---)
+      .trim();
+  
+    return plainText;
+  };
diff --git a/extractor/src/utils/convertToZodSchema.d.ts b/extractor/src/utils/convertToZodSchema.d.ts
new file mode 100644
index 0000000..175b13d
--- /dev/null
+++ b/extractor/src/utils/convertToZodSchema.d.ts
@@ -0,0 +1,15 @@
+import { z, ZodTypeAny } from 'zod';
+export interface SchemaField {
+    name: string;
+    type: 'string' | 'number' | 'boolean' | 'enum' | 'object' | 'array';
+    description?: string;
+    values?: string[];
+    children?: SchemaField[];
+}
+/**
+ * Converts an array of field definitions into a Zod schema.
+ * @param object - Array of field definitions.
+ * @returns - A Zod object schema.
+ */
+export declare const convertToZodSchema: (object: SchemaField[]) => z.ZodObject<Record<string, ZodTypeAny>>;
+//# sourceMappingURL=convertToZodSchema.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/utils/convertToZodSchema.d.ts.map b/extractor/src/utils/convertToZodSchema.d.ts.map
new file mode 100644
index 0000000..cfa4634
--- /dev/null
+++ b/extractor/src/utils/convertToZodSchema.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"convertToZodSchema.d.ts","sourceRoot":"","sources":["convertToZodSchema.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,CAAC,EAAE,UAAU,EAAE,MAAM,KAAK,CAAC;AAEpC,MAAM,WAAW,WAAW;IAC1B,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,QAAQ,GAAG,QAAQ,GAAG,SAAS,GAAG,MAAM,GAAG,QAAQ,GAAG,OAAO,CAAC;IACpE,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,MAAM,CAAC,EAAE,MAAM,EAAE,CAAC;IAClB,QAAQ,CAAC,EAAE,WAAW,EAAE,CAAC;CAC1B;AAED;;;;GAIG;AACH,eAAO,MAAM,kBAAkB,WAAY,WAAW,EAAE,KAAG,CAAC,CAAC,SAAS,CAAC,MAAM,CAAC,MAAM,EAAE,UAAU,CAAC,CAsEhG,CAAC"}
\ No newline at end of file
diff --git a/extractor/src/utils/convertToZodSchema.js b/extractor/src/utils/convertToZodSchema.js
index 8f8f4f0..70a6e8d 100644
--- a/extractor/src/utils/convertToZodSchema.js
+++ b/extractor/src/utils/convertToZodSchema.js
@@ -1,78 +1,66 @@
 import { z } from 'zod';
-
 /**
  * Converts an array of field definitions into a Zod schema.
- * @param {Array} object - Array of field definitions.
- * @returns {ZodObject} - A Zod object schema.
+ * @param object - Array of field definitions.
+ * @returns - A Zod object schema.
  */
 export const convertToZodSchema = (object) => {
-  const createZodSchema = (fields) => {
-    const schema = {};
-
-    fields.forEach((item) => {
-      let zodType;
-
-      switch (item.type) {
-        case 'string':
-          zodType = z.string();
-          break;
-
-        case 'number':
-          zodType = z.number();
-          break;
-
-        case 'boolean':
-          zodType = z.boolean();
-          break;
-
-        case 'enum':
-          if (item.values && Array.isArray(item.values)) {
-            zodType = z.enum(item.values);
-          } else {
-            throw new Error(`Invalid "enum" type definition for ${item.name}`);
-          }
-          break;
-
-        case 'object':
-          if (item.children) {
-            zodType = z.object(createZodSchema(item.children));
-          } else {
-            throw new Error(`Invalid "object" type definition for ${item.name}`);
-          }
-          break;
-
-        case 'array':
-          if (item.children && item.children.length > 0) {
-            if (item.children.length === 1) {
-              const singleChild = item.children[0];
-              const childSchema = createZodSchema([singleChild]);
-              const childType = childSchema[singleChild.name];
-              
-              zodType = z.array(childType);
-            } else {
-              const arraySchema = createZodSchema(item.children);
-              zodType = z.array(z.object(arraySchema));
+    const createZodSchema = (fields) => {
+        const schema = {};
+        fields.forEach((item) => {
+            let zodType;
+            switch (item.type) {
+                case 'string':
+                    zodType = z.string();
+                    break;
+                case 'number':
+                    zodType = z.number();
+                    break;
+                case 'boolean':
+                    zodType = z.boolean();
+                    break;
+                case 'enum':
+                    if (item.values && Array.isArray(item.values) && item.values.length > 0) {
+                        zodType = z.enum(item.values);
+                    }
+                    else {
+                        throw new Error(`Invalid "enum" type definition for ${item.name}`);
+                    }
+                    break;
+                case 'object':
+                    if (item.children) {
+                        zodType = z.object(createZodSchema(item.children));
+                    }
+                    else {
+                        throw new Error(`Invalid "object" type definition for ${item.name}`);
+                    }
+                    break;
+                case 'array':
+                    if (item.children && item.children.length > 0) {
+                        if (item.children.length === 1) {
+                            const singleChild = item.children[0];
+                            const childSchema = createZodSchema([singleChild]);
+                            const childType = childSchema[singleChild.name];
+                            zodType = z.array(childType);
+                        }
+                        else {
+                            const arraySchema = createZodSchema(item.children);
+                            zodType = z.array(z.object(arraySchema));
+                        }
+                    }
+                    else {
+                        throw new Error(`Invalid or unsupported "array" type definition for ${item.name}`);
+                    }
+                    break;
+                default:
+                    throw new Error(`Unsupported type "${item.type}" for field ${item.name}`);
             }
-          } else {
-            throw new Error(
-              `Invalid or unsupported "array" type definition for ${item.name}`
-            );
-          }
-          break;
-
-        default:
-          throw new Error(`Unsupported type "${item.type}" for field ${item.name}`);
-      }
-
-      if (item.description) {
-        zodType = zodType.describe(item.description);
-      }
-
-      schema[item.name] = zodType;
-    });
-
-    return schema;
-  };
-
-  return z.object(createZodSchema(object));
+            if (item.description) {
+                zodType = zodType.describe(item.description);
+            }
+            schema[item.name] = zodType;
+        });
+        return schema;
+    };
+    return z.object(createZodSchema(object));
 };
diff --git a/extractor/src/utils/convertToZodSchema.ts b/extractor/src/utils/convertToZodSchema.ts
new file mode 100644
index 0000000..b77e9be
--- /dev/null
+++ b/extractor/src/utils/convertToZodSchema.ts
@@ -0,0 +1,86 @@
+import { z, ZodTypeAny } from 'zod';
+
+export interface SchemaField {
+  name: string;
+  type: 'string' | 'number' | 'boolean' | 'enum' | 'object' | 'array';
+  description?: string;
+  values?: string[];
+  children?: SchemaField[];
+}
+
+/**
+ * Converts an array of field definitions into a Zod schema.
+ * @param object - Array of field definitions.
+ * @returns - A Zod object schema.
+ */
+export const convertToZodSchema = (object: SchemaField[]): z.ZodObject<Record<string, ZodTypeAny>> => {
+  const createZodSchema = (fields: SchemaField[]): Record<string, ZodTypeAny> => {
+    const schema: Record<string, ZodTypeAny> = {};
+
+    fields.forEach((item) => {
+      let zodType: ZodTypeAny;
+
+      switch (item.type) {
+        case 'string':
+          zodType = z.string();
+          break;
+
+        case 'number':
+          zodType = z.number();
+          break;
+
+        case 'boolean':
+          zodType = z.boolean();
+          break;
+
+        case 'enum':
+          if (item.values && Array.isArray(item.values) && item.values.length > 0) {
+            zodType = z.enum(item.values as [string, ...string[]]);
+          } else {
+            throw new Error(`Invalid "enum" type definition for ${item.name}`);
+          }
+          break;
+
+        case 'object':
+          if (item.children) {
+            zodType = z.object(createZodSchema(item.children));
+          } else {
+            throw new Error(`Invalid "object" type definition for ${item.name}`);
+          }
+          break;
+
+        case 'array':
+          if (item.children && item.children.length > 0) {
+            if (item.children.length === 1) {
+              const singleChild = item.children[0];
+              const childSchema = createZodSchema([singleChild]);
+              const childType = childSchema[singleChild.name];
+              
+              zodType = z.array(childType);
+            } else {
+              const arraySchema = createZodSchema(item.children);
+              zodType = z.array(z.object(arraySchema));
+            }
+          } else {
+            throw new Error(
+              `Invalid or unsupported "array" type definition for ${item.name}`
+            );
+          }
+          break;
+
+        default:
+          throw new Error(`Unsupported type "${item.type}" for field ${item.name}`);
+      }
+
+      if (item.description) {
+        zodType = zodType.describe(item.description);
+      }
+
+      schema[item.name] = zodType;
+    });
+
+    return schema;
+  };
+
+  return z.object(createZodSchema(object));
+};
diff --git a/extractor/src/utils/fileValidator.d.ts b/extractor/src/utils/fileValidator.d.ts
new file mode 100644
index 0000000..efeb781
--- /dev/null
+++ b/extractor/src/utils/fileValidator.d.ts
@@ -0,0 +1,7 @@
+/**
+ * Function to check if a file is valid based on its URL or MIME type
+ * @param file - The URL to the file
+ * @returns - Resolves to true if the file is valid, false otherwise
+ */
+export declare function isValidFile(file: string): Promise<boolean>;
+//# sourceMappingURL=fileValidator.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/utils/fileValidator.d.ts.map b/extractor/src/utils/fileValidator.d.ts.map
new file mode 100644
index 0000000..886cd87
--- /dev/null
+++ b/extractor/src/utils/fileValidator.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"fileValidator.d.ts","sourceRoot":"","sources":["fileValidator.ts"],"names":[],"mappings":"AAEA;;;;GAIG;AACH,wBAAsB,WAAW,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,OAAO,CAAC,CA4BhE"}
\ No newline at end of file
diff --git a/extractor/src/utils/fileValidator.js b/extractor/src/utils/fileValidator.js
index 65fd347..d311027 100644
--- a/extractor/src/utils/fileValidator.js
+++ b/extractor/src/utils/fileValidator.js
@@ -1,9 +1,8 @@
 import axios from 'axios';
-
 /**
  * Function to check if a file is valid based on its URL or MIME type
- * @param {string} file - The URL to the file
- * @returns {Promise<boolean>} - Resolves to true if the file is valid, false otherwise
+ * @param file - The URL to the file
+ * @returns - Resolves to true if the file is valid, false otherwise
  */
 export async function isValidFile(file) {
     const allowedExtensions = ['pdf', 'png', 'jpg', 'jpeg', 'txt', 'docx', 'html'];
@@ -16,20 +15,18 @@ export async function isValidFile(file) {
         docx: 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
         html: 'text/html',
     };
-
     const urlPath = new URL(file).pathname;
     const extensionRegex = new RegExp(`\\.(${allowedExtensions.join('|')})$`, 'i');
-
     if (!extensionRegex.test(urlPath)) {
         return false;
     }
-
     // Optional: Check the MIME type if query parameters are used
     try {
         const response = await axios.head(file);
         const contentType = response.headers['content-type'];
         return Object.values(allowedMimeTypes).some(mime => contentType.startsWith(mime));
-    } catch (error) {
+    }
+    catch (error) {
         console.error('Error checking MIME type:', error);
         return false;
     }
diff --git a/extractor/src/utils/fileValidator.ts b/extractor/src/utils/fileValidator.ts
new file mode 100644
index 0000000..83b6945
--- /dev/null
+++ b/extractor/src/utils/fileValidator.ts
@@ -0,0 +1,57 @@
+import axios from 'axios';
+import path from 'path';
+import fs from 'fs-extra';
+
+/**
+ * Function to check if a file is valid based on its URL or file path
+ * @param file - The URL or file path to the file
+ * @returns - Resolves to true if the file is valid, false otherwise
+ */
+export async function isValidFile(file: string): Promise<boolean> {
+    const allowedExtensions = ['pdf', 'png', 'jpg', 'jpeg', 'txt', 'docx', 'html'];
+    const allowedMimeTypes: Record<string, string> = {
+        pdf: 'application/pdf',
+        png: 'image/png',
+        jpg: 'image/jpeg',
+        jpeg: 'image/jpeg',
+        txt: 'text/plain',
+        docx: 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
+        html: 'text/html',
+    };
+
+    // Check if it's a local file path
+    if (!file.startsWith('http://') && !file.startsWith('https://')) {
+        // Handle local file path
+        const extension = path.extname(file).substring(1).toLowerCase();
+        
+        if (!allowedExtensions.includes(extension)) {
+            return false;
+        }
+        
+        // Check if file exists
+        try {
+            await fs.access(file);
+            return true;
+        } catch {
+            return false;
+        }
+    }
+
+    // Handle URL
+    const urlPath = new URL(file).pathname;
+    const extensionRegex = new RegExp(`\\.(${allowedExtensions.join('|')})$`, 'i');
+
+    if (!extensionRegex.test(urlPath)) {
+        return false;
+    }
+
+    // Optional: Check the MIME type if query parameters are used
+    try {
+        const response = await axios.head(file);
+        const contentType = response.headers['content-type'];
+        return Object.values(allowedMimeTypes).some(mime => contentType.startsWith(mime));
+    } catch (error) {
+        console.error('Error checking MIME type:', error);
+        return false;
+    }
+}
diff --git a/extractor/src/utils/generateMarkdown.d.ts b/extractor/src/utils/generateMarkdown.d.ts
new file mode 100644
index 0000000..c70d6cd
--- /dev/null
+++ b/extractor/src/utils/generateMarkdown.d.ts
@@ -0,0 +1,8 @@
+interface Page {
+    content: string;
+    page: number;
+    contentLength: number;
+}
+export declare const generateMarkdownDocument: (pages: Page[]) => Promise<string>;
+export {};
+//# sourceMappingURL=generateMarkdown.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/utils/generateMarkdown.d.ts.map b/extractor/src/utils/generateMarkdown.d.ts.map
new file mode 100644
index 0000000..cd07449
--- /dev/null
+++ b/extractor/src/utils/generateMarkdown.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"generateMarkdown.d.ts","sourceRoot":"","sources":["generateMarkdown.ts"],"names":[],"mappings":"AAAA,UAAU,IAAI;IACZ,OAAO,EAAE,MAAM,CAAC;IAChB,IAAI,EAAE,MAAM,CAAC;IACb,aAAa,EAAE,MAAM,CAAC;CACvB;AAED,eAAO,MAAM,wBAAwB,UAAiB,IAAI,EAAE,KAAG,OAAO,CAAC,MAAM,CAW5E,CAAC"}
\ No newline at end of file
diff --git a/extractor/src/utils/generateMarkdown.js b/extractor/src/utils/generateMarkdown.js
index 91a9a84..a32a148 100644
--- a/extractor/src/utils/generateMarkdown.js
+++ b/extractor/src/utils/generateMarkdown.js
@@ -1,12 +1,12 @@
 export const generateMarkdownDocument = async (pages) => {
-  try {
-    // Combine all markdown pages into a single string
-    const markdownContent = pages.map((page) => page.content).join("\n\n---\n\n");
-
-    // Return the combined markdown string directly
-    return markdownContent;
-  } catch (error) {
-    console.error('Error generating markdown:', error);
-    throw error;
-  }
+    try {
+        // Combine all markdown pages into a single string
+        const markdownContent = pages.map((page) => page.content).join("\n\n---\n\n");
+        // Return the combined markdown string directly
+        return markdownContent;
+    }
+    catch (error) {
+        console.error('Error generating markdown:', error);
+        throw error;
+    }
 };
diff --git a/extractor/src/utils/generateMarkdown.ts b/extractor/src/utils/generateMarkdown.ts
new file mode 100644
index 0000000..ba5eb78
--- /dev/null
+++ b/extractor/src/utils/generateMarkdown.ts
@@ -0,0 +1,18 @@
+interface Page {
+  content: string;
+  page: number;
+  contentLength: number;
+}
+
+export const generateMarkdownDocument = async (pages: Page[]): Promise<string> => {
+  try {
+    // Combine all markdown pages into a single string
+    const markdownContent = pages.map((page) => page.content).join("\n\n---\n\n");
+
+    // Return the combined markdown string directly
+    return markdownContent;
+  } catch (error) {
+    console.error('Error generating markdown:', error);
+    throw error;
+  }
+};
diff --git a/extractor/src/utils/pdfValidator.d.ts b/extractor/src/utils/pdfValidator.d.ts
new file mode 100644
index 0000000..92c6319
--- /dev/null
+++ b/extractor/src/utils/pdfValidator.d.ts
@@ -0,0 +1,7 @@
+/**
+ * Function to check if a file is a PDF based on its URL or MIME type
+ * @param file - The URL to the file
+ * @returns - Resolves to true if the file is a PDF, false otherwise
+ */
+export declare function isPdfFile(file: string): Promise<boolean>;
+//# sourceMappingURL=pdfValidator.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/utils/pdfValidator.d.ts.map b/extractor/src/utils/pdfValidator.d.ts.map
new file mode 100644
index 0000000..ef2e75d
--- /dev/null
+++ b/extractor/src/utils/pdfValidator.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"pdfValidator.d.ts","sourceRoot":"","sources":["pdfValidator.ts"],"names":[],"mappings":"AAEA;;;;GAIG;AACH,wBAAsB,SAAS,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,OAAO,CAAC,CAgB9D"}
\ No newline at end of file
diff --git a/extractor/src/utils/pdfValidator.js b/extractor/src/utils/pdfValidator.js
index f1dc67e..4d13190 100644
--- a/extractor/src/utils/pdfValidator.js
+++ b/extractor/src/utils/pdfValidator.js
@@ -1,9 +1,8 @@
 import axios from 'axios';
-
 /**
  * Function to check if a file is a PDF based on its URL or MIME type
- * @param {string} file - The URL to the file
- * @returns {Promise<boolean>} - Resolves to true if the file is a PDF, false otherwise
+ * @param file - The URL to the file
+ * @returns - Resolves to true if the file is a PDF, false otherwise
  */
 export async function isPdfFile(file) {
     const urlPath = new URL(file).pathname;
@@ -11,13 +10,13 @@ export async function isPdfFile(file) {
     if (pdfExtensionRegex.test(urlPath)) {
         return true;
     }
-
     // Optional: Check the MIME type if query parameters are used
     try {
         const response = await axios.head(file);
         const contentType = response.headers['content-type'];
         return contentType === 'application/pdf';
-    } catch (error) {
+    }
+    catch (error) {
         console.error('Error checking MIME type:', error);
         return false;
     }
diff --git a/extractor/src/utils/pdfValidator.ts b/extractor/src/utils/pdfValidator.ts
new file mode 100644
index 0000000..9d65cf4
--- /dev/null
+++ b/extractor/src/utils/pdfValidator.ts
@@ -0,0 +1,24 @@
+import axios from 'axios';
+
+/**
+ * Function to check if a file is a PDF based on its URL or MIME type
+ * @param file - The URL to the file
+ * @returns - Resolves to true if the file is a PDF, false otherwise
+ */
+export async function isPdfFile(file: string): Promise<boolean> {
+    const urlPath = new URL(file).pathname;
+    const pdfExtensionRegex = /\.pdf$/i;
+    if (pdfExtensionRegex.test(urlPath)) {
+        return true;
+    }
+
+    // Optional: Check the MIME type if query parameters are used
+    try {
+        const response = await axios.head(file);
+        const contentType = response.headers['content-type'];
+        return contentType === 'application/pdf';
+    } catch (error) {
+        console.error('Error checking MIME type:', error);
+        return false;
+    }
+}
diff --git a/extractor/src/utils/schemaValidator.d.ts b/extractor/src/utils/schemaValidator.d.ts
new file mode 100644
index 0000000..7a507fe
--- /dev/null
+++ b/extractor/src/utils/schemaValidator.d.ts
@@ -0,0 +1,13 @@
+import { SchemaField } from './convertToZodSchema.js';
+interface ValidationResult {
+    isValid: boolean;
+    errors: string[];
+}
+/**
+ * Validates the schema format to ensure it meets the required structure.
+ * @param schema - The schema to validate.
+ * @returns - { isValid: boolean, errors: Array<string> }
+ */
+export declare function validateSchema(schema: SchemaField[]): ValidationResult;
+export {};
+//# sourceMappingURL=schemaValidator.d.ts.map
\ No newline at end of file
diff --git a/extractor/src/utils/schemaValidator.d.ts.map b/extractor/src/utils/schemaValidator.d.ts.map
new file mode 100644
index 0000000..62e2fe7
--- /dev/null
+++ b/extractor/src/utils/schemaValidator.d.ts.map
@@ -0,0 +1 @@
+{"version":3,"file":"schemaValidator.d.ts","sourceRoot":"","sources":["schemaValidator.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,WAAW,EAAE,MAAM,yBAAyB,CAAC;AAEtD,UAAU,gBAAgB;IACxB,OAAO,EAAE,OAAO,CAAC;IACjB,MAAM,EAAE,MAAM,EAAE,CAAC;CAClB;AAED;;;;GAIG;AACH,wBAAgB,cAAc,CAAC,MAAM,EAAE,WAAW,EAAE,GAAG,gBAAgB,CAyDtE"}
\ No newline at end of file
diff --git a/extractor/src/utils/schemaValidator.js b/extractor/src/utils/schemaValidator.js
index 5b7cd00..330fedb 100644
--- a/extractor/src/utils/schemaValidator.js
+++ b/extractor/src/utils/schemaValidator.js
@@ -1,64 +1,59 @@
 /**
  * Validates the schema format to ensure it meets the required structure.
- * @param {Array} schema - The schema to validate.
- * @returns {Object} - { isValid: boolean, errors: Array<string> }
+ * @param schema - The schema to validate.
+ * @returns - { isValid: boolean, errors: Array<string> }
  */
-
 export function validateSchema(schema) {
-  const validTypes = ["string", "number", "array", "object", "boolean", "enum"];
-  let errors = [];
-
-  if (!Array.isArray(schema)) {
-    errors.push("Schema must be an array.");
-    return { isValid: false, errors };
-  }
-
-  function validateField(field, path) {
-    if (!field.hasOwnProperty("name") || typeof field.name !== "string" || field.name.trim() === "") {
-      errors.push(`"name" is required and should be a non-empty string at ${path}`);
+    const validTypes = ["string", "number", "array", "object", "boolean", "enum"];
+    let errors = [];
+    if (!Array.isArray(schema)) {
+        errors.push("Schema must be an array.");
+        return { isValid: false, errors };
     }
-
-    if (!field.hasOwnProperty("type") || !validTypes.includes(field.type)) {
-      errors.push(`"type" is required and must be one of ${validTypes.join(", ")} at ${path}`);
+    function validateField(field, path) {
+        if (!field.hasOwnProperty("name") || typeof field.name !== "string" || field.name.trim() === "") {
+            errors.push(`"name" is required and should be a non-empty string at ${path}`);
+        }
+        if (!field.hasOwnProperty("type") || !validTypes.includes(field.type)) {
+            errors.push(`"type" is required and must be one of ${validTypes.join(", ")} at ${path}`);
+        }
+        if (!field.hasOwnProperty("description") || typeof field.description !== "string" || field.description.trim() === "") {
+            errors.push(`"description" is required and should be a non-empty string at ${path}`);
+        }
+        if (field.type === "enum") {
+            if (!field.hasOwnProperty("values") || !Array.isArray(field.values) || field.values.length === 0) {
+                errors.push(`"values" is required and must be a non-empty array for enum "${field.name}" at ${path}`);
+            }
+            else if (!field.values.every((value) => typeof value === "string")) {
+                errors.push(`"values" for enum "${field.name}" at ${path} must be an array of strings`);
+            }
+        }
+        if (field.type === "array") {
+            if (!field.hasOwnProperty("children")) {
+                errors.push(`"children" property is required for array "${field.name}" at ${path}`);
+            }
+            else if (!Array.isArray(field.children) || field.children.length === 0) {
+                errors.push(`"children" must be a non-empty array for "${field.name}" at ${path}`);
+            }
+            else {
+                field.children.forEach((child, index) => validateField(child, `${path}.children[${index}]`));
+            }
+        }
+        if (field.type === "object") {
+            if (!field.hasOwnProperty("children")) {
+                errors.push(`"children" property is required for object "${field.name}" at ${path}`);
+            }
+            else if (!Array.isArray(field.children) || field.children.length === 0) {
+                errors.push(`"children" must be a non-empty array for "${field.name}" at ${path}`);
+            }
+            else {
+                field.children.forEach((child, index) => validateField(child, `${path}.children[${index}]`));
+            }
+        }
     }
-
-    if (!field.hasOwnProperty("description") || typeof field.description !== "string" || field.description.trim() === "") {
-      errors.push(`"description" is required and should be a non-empty string at ${path}`);
-    }
-
-    if (field.type === "enum") {
-      if (!field.hasOwnProperty("values") || !Array.isArray(field.values) || field.values.length === 0) {
-        errors.push(`"values" is required and must be a non-empty array for enum "${field.name}" at ${path}`);
-      } else if (!field.values.every((value) => typeof value === "string")) {
-        errors.push(`"values" for enum "${field.name}" at ${path} must be an array of strings`);
-      }
-    }
-
-    if (field.type === "array") {
-      if (!field.hasOwnProperty("children")) {
-        errors.push(`"children" property is required for array "${field.name}" at ${path}`);
-      } else if (!Array.isArray(field.children) || field.children.length === 0) {
-        errors.push(`"children" must be a non-empty array for "${field.name}" at ${path}`);
-      } else {
-        field.children.forEach((child, index) => validateField(child, `${path}.children[${index}]`));
-      }
-    }
-
-    if (field.type === "object") {
-      if (!field.hasOwnProperty("children")) {
-        errors.push(`"children" property is required for object "${field.name}" at ${path}`);
-      } else if (!Array.isArray(field.children) || field.children.length === 0) {
-        errors.push(`"children" must be a non-empty array for "${field.name}" at ${path}`);
-      } else {
-        field.children.forEach((child, index) => validateField(child, `${path}.children[${index}]`));
-      }
-    }
-  }
-
-  schema.forEach((field, index) => validateField(field, `schema[${index}]`));
-
-  return {
-    isValid: errors.length === 0,
-    errors,
-  };
-}  
\ No newline at end of file
+    schema.forEach((field, index) => validateField(field, `schema[${index}]`));
+    return {
+        isValid: errors.length === 0,
+        errors,
+    };
+}
diff --git a/extractor/src/utils/schemaValidator.ts b/extractor/src/utils/schemaValidator.ts
new file mode 100644
index 0000000..3dd462a
--- /dev/null
+++ b/extractor/src/utils/schemaValidator.ts
@@ -0,0 +1,70 @@
+import { SchemaField } from './convertToZodSchema.js';
+
+interface ValidationResult {
+  isValid: boolean;
+  errors: string[];
+}
+
+/**
+ * Validates the schema format to ensure it meets the required structure.
+ * @param schema - The schema to validate.
+ * @returns - { isValid: boolean, errors: Array<string> }
+ */
+export function validateSchema(schema: SchemaField[]): ValidationResult {
+  const validTypes = ["string", "number", "array", "object", "boolean", "enum"];
+  let errors: string[] = [];
+
+  if (!Array.isArray(schema)) {
+    errors.push("Schema must be an array.");
+    return { isValid: false, errors };
+  }
+
+  function validateField(field: any, path: string): void {
+    if (!field.hasOwnProperty("name") || typeof field.name !== "string" || field.name.trim() === "") {
+      errors.push(`"name" is required and should be a non-empty string at ${path}`);
+    }
+
+    if (!field.hasOwnProperty("type") || !validTypes.includes(field.type)) {
+      errors.push(`"type" is required and must be one of ${validTypes.join(", ")} at ${path}`);
+    }
+
+    if (!field.hasOwnProperty("description") || typeof field.description !== "string" || field.description.trim() === "") {
+      errors.push(`"description" is required and should be a non-empty string at ${path}`);
+    }
+
+    if (field.type === "enum") {
+      if (!field.hasOwnProperty("values") || !Array.isArray(field.values) || field.values.length === 0) {
+        errors.push(`"values" is required and must be a non-empty array for enum "${field.name}" at ${path}`);
+      } else if (!field.values.every((value: any) => typeof value === "string")) {
+        errors.push(`"values" for enum "${field.name}" at ${path} must be an array of strings`);
+      }
+    }
+
+    if (field.type === "array") {
+      if (!field.hasOwnProperty("children")) {
+        errors.push(`"children" property is required for array "${field.name}" at ${path}`);
+      } else if (!Array.isArray(field.children) || field.children.length === 0) {
+        errors.push(`"children" must be a non-empty array for "${field.name}" at ${path}`);
+      } else {
+        field.children.forEach((child: any, index: number) => validateField(child, `${path}.children[${index}]`));
+      }
+    }
+
+    if (field.type === "object") {
+      if (!field.hasOwnProperty("children")) {
+        errors.push(`"children" property is required for object "${field.name}" at ${path}`);
+      } else if (!Array.isArray(field.children) || field.children.length === 0) {
+        errors.push(`"children" must be a non-empty array for "${field.name}" at ${path}`);
+      } else {
+        field.children.forEach((child: any, index: number) => validateField(child, `${path}.children[${index}]`));
+      }
+    }
+  }
+
+  schema.forEach((field, index) => validateField(field, `schema[${index}]`));
+
+  return {
+    isValid: errors.length === 0,
+    errors,
+  };
+}
diff --git a/extractor/tsconfig.json b/extractor/tsconfig.json
new file mode 100644
index 0000000..42e9d4b
--- /dev/null
+++ b/extractor/tsconfig.json
@@ -0,0 +1,27 @@
+{
+  "compilerOptions": {
+    "target": "ES2020",
+    "module": "ES2020",
+    "moduleResolution": "node",
+    "declaration": true,
+    "declarationMap": true,
+    "outDir": "./dist",
+    "rootDir": "./src",
+    "strict": true,
+    "esModuleInterop": true,
+    "skipLibCheck": true,
+    "forceConsistentCasingInFileNames": true,
+    "resolveJsonModule": true,
+    "allowSyntheticDefaultImports": true,
+    "composite": true,
+    "baseUrl": ".",
+    "paths": {
+      "core": ["../core/src/index.ts"]
+    }
+  },
+  "include": ["src/**/*"],
+  "exclude": ["node_modules", "dist"],
+  "references": [
+    { "path": "../core" }
+  ]
+}
diff --git a/package-lock.json b/package-lock.json
index c5db782..4565ea4 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -10,10 +10,11 @@
       "license": "AGPL-3.0",
       "workspaces": [
         "core",
-        "extractor"
+        "extractor",
+        "cli"
       ],
       "dependencies": {
-        "axios": "^1.7.7",
+        "axios": "^1.12.2",
         "child_process": "^1.0.2",
         "core": "file:./core",
         "dotenv": "^16.4.5",
@@ -33,6 +34,41 @@
         "zod": "^3.23.8"
       }
     },
+    "cli": {
+      "version": "1.0.0",
+      "dependencies": {
+        "chalk": "^5.3.0",
+        "commander": "^12.0.0",
+        "dotenv": "^16.4.5",
+        "extractor": "*",
+        "ora": "^8.0.1"
+      },
+      "bin": {
+        "documind": "dist/index.js"
+      },
+      "devDependencies": {
+        "@types/node": "^20.14.11",
+        "tsc-alias": "^1.8.8",
+        "typescript": "^5.6.3"
+      }
+    },
+    "cli/node_modules/@types/node": {
+      "version": "20.19.20",
+      "resolved": "https://registry.npmjs.org/@types/node/-/node-20.19.20.tgz",
+      "integrity": "sha512-2Q7WS25j4pS1cS8yw3d6buNCVJukOTeQ39bAnwR6sOJbaxvyCGebzTMypDFN82CxBLnl+lSWVdCCWbRY6y9yZQ==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "undici-types": "~6.21.0"
+      }
+    },
+    "cli/node_modules/undici-types": {
+      "version": "6.21.0",
+      "resolved": "https://registry.npmjs.org/undici-types/-/undici-types-6.21.0.tgz",
+      "integrity": "sha512-iwDZqg0QAGrg9Rav5H4n0M64c3mkR59cJ6wQp+7C4nI0gsmExaedaYLNO44eT4AtBBwjbTiGPMlt2Md0T9H9JQ==",
+      "dev": true,
+      "license": "MIT"
+    },
     "core": {
       "version": "1.0.0",
       "dependencies": {
@@ -85,9 +121,30 @@
         "zod-to-json-schema": "^3.24.2"
       },
       "devDependencies": {
-        "nodemon": "^3.1.7"
+        "@types/node": "^20.14.11",
+        "@types/uuid": "^11.0.0",
+        "nodemon": "^3.1.7",
+        "tsc-alias": "^1.8.8",
+        "typescript": "^5.6.3"
+      }
+    },
+    "extractor/node_modules/@types/node": {
+      "version": "20.19.20",
+      "resolved": "https://registry.npmjs.org/@types/node/-/node-20.19.20.tgz",
+      "integrity": "sha512-2Q7WS25j4pS1cS8yw3d6buNCVJukOTeQ39bAnwR6sOJbaxvyCGebzTMypDFN82CxBLnl+lSWVdCCWbRY6y9yZQ==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "undici-types": "~6.21.0"
       }
     },
+    "extractor/node_modules/undici-types": {
+      "version": "6.21.0",
+      "resolved": "https://registry.npmjs.org/undici-types/-/undici-types-6.21.0.tgz",
+      "integrity": "sha512-iwDZqg0QAGrg9Rav5H4n0M64c3mkR59cJ6wQp+7C4nI0gsmExaedaYLNO44eT4AtBBwjbTiGPMlt2Md0T9H9JQ==",
+      "dev": true,
+      "license": "MIT"
+    },
     "node_modules/@ai-sdk/google": {
       "version": "1.1.14",
       "resolved": "https://registry.npmjs.org/@ai-sdk/google/-/google-1.1.14.tgz",
@@ -575,6 +632,44 @@
         "url": "https://opencollective.com/libvips"
       }
     },
+    "node_modules/@nodelib/fs.scandir": {
+      "version": "2.1.5",
+      "resolved": "https://registry.npmjs.org/@nodelib/fs.scandir/-/fs.scandir-2.1.5.tgz",
+      "integrity": "sha512-vq24Bq3ym5HEQm2NKCr3yXDwjc7vTsEThRDnkp2DK9p1uqLR+DHurm/NOTo0KG7HYHU7eppKZj3MyqYuMBf62g==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "@nodelib/fs.stat": "2.0.5",
+        "run-parallel": "^1.1.9"
+      },
+      "engines": {
+        "node": ">= 8"
+      }
+    },
+    "node_modules/@nodelib/fs.stat": {
+      "version": "2.0.5",
+      "resolved": "https://registry.npmjs.org/@nodelib/fs.stat/-/fs.stat-2.0.5.tgz",
+      "integrity": "sha512-RkhPPp2zrqDAQA/2jNhnztcPAlv64XdhIp7a7454A5ovI7Bukxgt7MX7udwAu3zg1DcpPU0rz3VV1SeaqvY4+A==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">= 8"
+      }
+    },
+    "node_modules/@nodelib/fs.walk": {
+      "version": "1.2.8",
+      "resolved": "https://registry.npmjs.org/@nodelib/fs.walk/-/fs.walk-1.2.8.tgz",
+      "integrity": "sha512-oGB+UxlgWcgQkgwo8GcEGwemoTFt3FIO9ababBmaGwXIoBKZ+GTy0pP185beGg7Llih/NSHSV2XAs1lnznocSg==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "@nodelib/fs.scandir": "2.1.5",
+        "fastq": "^1.6.0"
+      },
+      "engines": {
+        "node": ">= 8"
+      }
+    },
     "node_modules/@opentelemetry/api": {
       "version": "1.9.0",
       "resolved": "https://registry.npmjs.org/@opentelemetry/api/-/api-1.9.0.tgz",
@@ -655,6 +750,17 @@
         "form-data": "^4.0.0"
       }
     },
+    "node_modules/@types/uuid": {
+      "version": "11.0.0",
+      "resolved": "https://registry.npmjs.org/@types/uuid/-/uuid-11.0.0.tgz",
+      "integrity": "sha512-HVyk8nj2m+jcFRNazzqyVKiZezyhDKrGUA3jlEcg/nZ6Ms+qHwocba1Y/AaVaznJTAM9xpdFSh+ptbNrhOGvZA==",
+      "deprecated": "This is a stub types definition. uuid provides its own type definitions, so you do not need this installed.",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "uuid": "*"
+      }
+    },
     "node_modules/abort-controller": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/abort-controller/-/abort-controller-3.0.0.tgz",
@@ -708,6 +814,18 @@
         }
       }
     },
+    "node_modules/ansi-regex": {
+      "version": "6.2.2",
+      "resolved": "https://registry.npmjs.org/ansi-regex/-/ansi-regex-6.2.2.tgz",
+      "integrity": "sha512-Bq3SmSpyFHaWjPk8If9yc6svM8c56dB5BAtW4Qbw5jHTwwXXcTLoRMkpDJp6VL0XzlWaCHTXrkFURMYmD0sLqg==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=12"
+      },
+      "funding": {
+        "url": "https://github.com/chalk/ansi-regex?sponsor=1"
+      }
+    },
     "node_modules/anymatch": {
       "version": "3.1.3",
       "resolved": "https://registry.npmjs.org/anymatch/-/anymatch-3.1.3.tgz",
@@ -734,6 +852,16 @@
       "integrity": "sha512-L0XlBwfx9QetHOsbLDrE/vh2t018w9462HM3iaFfxRiK83aJjAt/Ja3NMkOW7FICwWTlQBa3ZbL5FKhuQWkDrg==",
       "license": "MIT"
     },
+    "node_modules/array-union": {
+      "version": "2.1.0",
+      "resolved": "https://registry.npmjs.org/array-union/-/array-union-2.1.0.tgz",
+      "integrity": "sha512-HGyxoOTYUyCM6stUe6EJgnd4EoewAI7zMdfqO+kGjnlZmBDz/cR5pf8r/cR4Wq60sL/p0IkcjUEEPwS3GFrIyw==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">=8"
+      }
+    },
     "node_modules/async": {
       "version": "3.2.6",
       "resolved": "https://registry.npmjs.org/async/-/async-3.2.6.tgz",
@@ -762,13 +890,13 @@
       }
     },
     "node_modules/axios": {
-      "version": "1.9.0",
-      "resolved": "https://registry.npmjs.org/axios/-/axios-1.9.0.tgz",
-      "integrity": "sha512-re4CqKTJaURpzbLHtIi6XpDv20/CnpXOtjRY5/CU32L8gU8ek9UIivcfvSWvmKEngmVbrUtPpdDwWDWL7DNHvg==",
+      "version": "1.12.2",
+      "resolved": "https://registry.npmjs.org/axios/-/axios-1.12.2.tgz",
+      "integrity": "sha512-vMJzPewAlRyOgxV2dU0Cuz2O8zzzx9VYtbJOaBgXFeLc4IV/Eg50n4LowmehOOR61S8ZMpc2K5Sa7g6A4jfkUw==",
       "license": "MIT",
       "dependencies": {
         "follow-redirects": "^1.15.6",
-        "form-data": "^4.0.0",
+        "form-data": "^4.0.4",
         "proxy-from-env": "^1.1.0"
       }
     },
@@ -830,6 +958,19 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/call-bind-apply-helpers": {
+      "version": "1.0.2",
+      "resolved": "https://registry.npmjs.org/call-bind-apply-helpers/-/call-bind-apply-helpers-1.0.2.tgz",
+      "integrity": "sha512-Sp1ablJ0ivDkSzjcaJdxEunN5/XvksFJ2sMBFfq6x0ryhQV/2b/KwFe21cMpmHtPOSij8K99/wSfoEuTObmuMQ==",
+      "license": "MIT",
+      "dependencies": {
+        "es-errors": "^1.3.0",
+        "function-bind": "^1.1.2"
+      },
+      "engines": {
+        "node": ">= 0.4"
+      }
+    },
     "node_modules/chalk": {
       "version": "5.4.1",
       "resolved": "https://registry.npmjs.org/chalk/-/chalk-5.4.1.tgz",
@@ -873,6 +1014,37 @@
         "fsevents": "~2.3.2"
       }
     },
+    "node_modules/cli": {
+      "resolved": "cli",
+      "link": true
+    },
+    "node_modules/cli-cursor": {
+      "version": "5.0.0",
+      "resolved": "https://registry.npmjs.org/cli-cursor/-/cli-cursor-5.0.0.tgz",
+      "integrity": "sha512-aCj4O5wKyszjMmDT4tZj93kxyydN/K5zPWSCe6/0AV/AA1pqe5ZBIw0a2ZfPQV7lL5/yb5HsUreJ6UFAF1tEQw==",
+      "license": "MIT",
+      "dependencies": {
+        "restore-cursor": "^5.0.0"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
+    "node_modules/cli-spinners": {
+      "version": "2.9.2",
+      "resolved": "https://registry.npmjs.org/cli-spinners/-/cli-spinners-2.9.2.tgz",
+      "integrity": "sha512-ywqV+5MmyL4E7ybXgKys4DugZbX0FC6LnwrhjuykIjnK9k8OQacQ7axGKnjDXWNhns0xot3bZI5h55H8yo9cJg==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=6"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/color": {
       "version": "4.2.3",
       "resolved": "https://registry.npmjs.org/color/-/color-4.2.3.tgz",
@@ -926,6 +1098,15 @@
         "node": ">= 0.8"
       }
     },
+    "node_modules/commander": {
+      "version": "12.1.0",
+      "resolved": "https://registry.npmjs.org/commander/-/commander-12.1.0.tgz",
+      "integrity": "sha512-Vw8qHK3bZM9y/P10u3Vib8o/DdkvA2OtPtZvD871QKjy74Wj1WSKFILMPRPSdUSx5RFK1arlJzEtA4PkFgnbuA==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      }
+    },
     "node_modules/concat-map": {
       "version": "0.0.1",
       "resolved": "https://registry.npmjs.org/concat-map/-/concat-map-0.0.1.tgz",
@@ -1006,6 +1187,19 @@
       "integrity": "sha512-IayShXAgj/QMXgB0IWmKx+rOPuGMhqm5w6jvFxmVenXKIzRqTAAsbBPT3kWQeGANj3jGgvcvv4yK6SxqYmikgw==",
       "license": "Apache-2.0"
     },
+    "node_modules/dir-glob": {
+      "version": "3.0.1",
+      "resolved": "https://registry.npmjs.org/dir-glob/-/dir-glob-3.0.1.tgz",
+      "integrity": "sha512-WkrWp9GR4KXfKGYzOLmTuGVi1UWFfws377n9cc55/tb6DuqyF6pcQ5AbiHEshaDpY9v6oaSr2XCDidGmMwdzIA==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "path-type": "^4.0.0"
+      },
+      "engines": {
+        "node": ">=8"
+      }
+    },
     "node_modules/dotenv": {
       "version": "16.4.5",
       "resolved": "https://registry.npmjs.org/dotenv/-/dotenv-16.4.5.tgz",
@@ -1018,18 +1212,35 @@
         "url": "https://dotenvx.com"
       }
     },
-    "node_modules/es-define-property": {
-      "version": "1.0.0",
-      "resolved": "https://registry.npmjs.org/es-define-property/-/es-define-property-1.0.0.tgz",
-      "integrity": "sha512-jxayLKShrEqqzJ0eumQbVhTYQM27CfT1T35+gCgDFoL82JLsXqTJ76zv6A0YLOgEnLUMvLzsDsGIrl8NFpT2gQ==",
+    "node_modules/dunder-proto": {
+      "version": "1.0.1",
+      "resolved": "https://registry.npmjs.org/dunder-proto/-/dunder-proto-1.0.1.tgz",
+      "integrity": "sha512-KIN/nDJBQRcXw0MLVhZE9iQHmG68qAVIBg9CqmUYjmQIhgij9U5MFvrqkUL5FbtyyzZuOeOt0zdeRe4UY7ct+A==",
       "license": "MIT",
       "dependencies": {
-        "get-intrinsic": "^1.2.4"
+        "call-bind-apply-helpers": "^1.0.1",
+        "es-errors": "^1.3.0",
+        "gopd": "^1.2.0"
       },
       "engines": {
         "node": ">= 0.4"
       }
     },
+    "node_modules/emoji-regex": {
+      "version": "10.5.0",
+      "resolved": "https://registry.npmjs.org/emoji-regex/-/emoji-regex-10.5.0.tgz",
+      "integrity": "sha512-lb49vf1Xzfx080OKA0o6l8DQQpV+6Vg95zyCJX9VB/BqKYlhG7N4wgROUUHRA+ZPUefLnteQOad7z1kT2bV7bg==",
+      "license": "MIT"
+    },
+    "node_modules/es-define-property": {
+      "version": "1.0.1",
+      "resolved": "https://registry.npmjs.org/es-define-property/-/es-define-property-1.0.1.tgz",
+      "integrity": "sha512-e3nRfgfUZ4rNGL232gUgX06QNyyez04KdjFrF+LTRoOXmrOgFKDg4BCdsjW8EnT69eqdYGmRpJwiPVYNrCaW3g==",
+      "license": "MIT",
+      "engines": {
+        "node": ">= 0.4"
+      }
+    },
     "node_modules/es-errors": {
       "version": "1.3.0",
       "resolved": "https://registry.npmjs.org/es-errors/-/es-errors-1.3.0.tgz",
@@ -1039,6 +1250,33 @@
         "node": ">= 0.4"
       }
     },
+    "node_modules/es-object-atoms": {
+      "version": "1.1.1",
+      "resolved": "https://registry.npmjs.org/es-object-atoms/-/es-object-atoms-1.1.1.tgz",
+      "integrity": "sha512-FGgH2h8zKNim9ljj7dankFPcICIK9Cp5bm+c2gQSYePhpaG5+esrLODihIorn+Pe6FGJzWhXQotPv73jTaldXA==",
+      "license": "MIT",
+      "dependencies": {
+        "es-errors": "^1.3.0"
+      },
+      "engines": {
+        "node": ">= 0.4"
+      }
+    },
+    "node_modules/es-set-tostringtag": {
+      "version": "2.1.0",
+      "resolved": "https://registry.npmjs.org/es-set-tostringtag/-/es-set-tostringtag-2.1.0.tgz",
+      "integrity": "sha512-j6vWzfrGVfyXxge+O0x5sh6cvxAog0a/4Rdd2K36zCMV5eJ+/+tOAngRO8cODMNWbVRdVlmGZQL2YS3yR8bIUA==",
+      "license": "MIT",
+      "dependencies": {
+        "es-errors": "^1.3.0",
+        "get-intrinsic": "^1.2.6",
+        "has-tostringtag": "^1.0.2",
+        "hasown": "^2.0.2"
+      },
+      "engines": {
+        "node": ">= 0.4"
+      }
+    },
     "node_modules/event-target-shim": {
       "version": "5.0.1",
       "resolved": "https://registry.npmjs.org/event-target-shim/-/event-target-shim-5.0.1.tgz",
@@ -1061,6 +1299,33 @@
       "resolved": "extractor",
       "link": true
     },
+    "node_modules/fast-glob": {
+      "version": "3.3.3",
+      "resolved": "https://registry.npmjs.org/fast-glob/-/fast-glob-3.3.3.tgz",
+      "integrity": "sha512-7MptL8U0cqcFdzIzwOTHoilX9x5BrNqye7Z/LuC7kCMRio1EMSyqRK3BEAUD7sXRq4iT4AzTVuZdhgQ2TCvYLg==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "@nodelib/fs.stat": "^2.0.2",
+        "@nodelib/fs.walk": "^1.2.3",
+        "glob-parent": "^5.1.2",
+        "merge2": "^1.3.0",
+        "micromatch": "^4.0.8"
+      },
+      "engines": {
+        "node": ">=8.6.0"
+      }
+    },
+    "node_modules/fastq": {
+      "version": "1.19.1",
+      "resolved": "https://registry.npmjs.org/fastq/-/fastq-1.19.1.tgz",
+      "integrity": "sha512-GwLTyxkCXjXbxqIhTsMI2Nui8huMPtnxg7krajPJAjnEG/iiOS7i+zCtWGZR9G0NBKbXKh6X9m9UIsYX/N6vvQ==",
+      "dev": true,
+      "license": "ISC",
+      "dependencies": {
+        "reusify": "^1.0.4"
+      }
+    },
     "node_modules/fill-range": {
       "version": "7.1.1",
       "resolved": "https://registry.npmjs.org/fill-range/-/fill-range-7.1.1.tgz",
@@ -1104,13 +1369,15 @@
       }
     },
     "node_modules/form-data": {
-      "version": "4.0.1",
-      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.1.tgz",
-      "integrity": "sha512-tzN8e4TX8+kkxGPK8D5u0FNmjPUjw3lwC9lSLxxoB/+GtsJG91CO8bSWy73APlgAZzZbXEYZJuxjkHH2w+Ezhw==",
+      "version": "4.0.4",
+      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.4.tgz",
+      "integrity": "sha512-KrGhL9Q4zjj0kiUt5OO4Mr/A/jlI2jDYs5eHBpYHPcBEVSiipAvn2Ko2HnPe20rmcuuvMHNdZFp+4IlGTMF0Ow==",
       "license": "MIT",
       "dependencies": {
         "asynckit": "^0.4.0",
         "combined-stream": "^1.0.8",
+        "es-set-tostringtag": "^2.1.0",
+        "hasown": "^2.0.2",
         "mime-types": "^2.1.12"
       },
       "engines": {
@@ -1174,17 +1441,34 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/get-east-asian-width": {
+      "version": "1.4.0",
+      "resolved": "https://registry.npmjs.org/get-east-asian-width/-/get-east-asian-width-1.4.0.tgz",
+      "integrity": "sha512-QZjmEOC+IT1uk6Rx0sX22V6uHWVwbdbxf1faPqJ1QhLdGgsRGCZoyaQBm/piRdJy/D2um6hM1UP7ZEeQ4EkP+Q==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/get-intrinsic": {
-      "version": "1.2.4",
-      "resolved": "https://registry.npmjs.org/get-intrinsic/-/get-intrinsic-1.2.4.tgz",
-      "integrity": "sha512-5uYhsJH8VJBTv7oslg4BznJYhDoRI6waYCxMmCdnTrcCrHA/fCFKoTFz2JKKE0HdDFUF7/oQuhzumXJK7paBRQ==",
+      "version": "1.3.0",
+      "resolved": "https://registry.npmjs.org/get-intrinsic/-/get-intrinsic-1.3.0.tgz",
+      "integrity": "sha512-9fSjSaos/fRIVIp+xSJlE6lfwhES7LNtKaCBIamHsjr2na1BiABJPo0mOjjz8GJDURarmCPGqaiVg5mfjb98CQ==",
       "license": "MIT",
       "dependencies": {
+        "call-bind-apply-helpers": "^1.0.2",
+        "es-define-property": "^1.0.1",
         "es-errors": "^1.3.0",
+        "es-object-atoms": "^1.1.1",
         "function-bind": "^1.1.2",
-        "has-proto": "^1.0.1",
-        "has-symbols": "^1.0.3",
-        "hasown": "^2.0.0"
+        "get-proto": "^1.0.1",
+        "gopd": "^1.2.0",
+        "has-symbols": "^1.1.0",
+        "hasown": "^2.0.2",
+        "math-intrinsics": "^1.1.0"
       },
       "engines": {
         "node": ">= 0.4"
@@ -1193,6 +1477,32 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/get-proto": {
+      "version": "1.0.1",
+      "resolved": "https://registry.npmjs.org/get-proto/-/get-proto-1.0.1.tgz",
+      "integrity": "sha512-sTSfBjoXBp89JvIKIefqw7U2CCebsc74kiY6awiGogKtoSGbgjYE/G/+l9sF3MWFPNc9IcoOC4ODfKHfxFmp0g==",
+      "license": "MIT",
+      "dependencies": {
+        "dunder-proto": "^1.0.1",
+        "es-object-atoms": "^1.0.0"
+      },
+      "engines": {
+        "node": ">= 0.4"
+      }
+    },
+    "node_modules/get-tsconfig": {
+      "version": "4.12.0",
+      "resolved": "https://registry.npmjs.org/get-tsconfig/-/get-tsconfig-4.12.0.tgz",
+      "integrity": "sha512-LScr2aNr2FbjAjZh2C6X6BxRx1/x+aTDExct/xyq2XKbYOiG5c0aK7pMsSuyc0brz3ibr/lbQiHD9jzt4lccJw==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "resolve-pkg-maps": "^1.0.0"
+      },
+      "funding": {
+        "url": "https://github.com/privatenumber/get-tsconfig?sponsor=1"
+      }
+    },
     "node_modules/glob-parent": {
       "version": "5.1.2",
       "resolved": "https://registry.npmjs.org/glob-parent/-/glob-parent-5.1.2.tgz",
@@ -1206,6 +1516,27 @@
         "node": ">= 6"
       }
     },
+    "node_modules/globby": {
+      "version": "11.1.0",
+      "resolved": "https://registry.npmjs.org/globby/-/globby-11.1.0.tgz",
+      "integrity": "sha512-jhIXaOzy1sb8IyocaruWSn1TjmnBVs8Ayhcy83rmxNJ8q2uWKCAj3CnJY+KpGSXCueAPc0i05kVvVKtP1t9S3g==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "array-union": "^2.1.0",
+        "dir-glob": "^3.0.1",
+        "fast-glob": "^3.2.9",
+        "ignore": "^5.2.0",
+        "merge2": "^1.4.1",
+        "slash": "^3.0.0"
+      },
+      "engines": {
+        "node": ">=10"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/gm": {
       "version": "1.25.0",
       "resolved": "https://registry.npmjs.org/gm/-/gm-1.25.0.tgz",
@@ -1222,12 +1553,12 @@
       }
     },
     "node_modules/gopd": {
-      "version": "1.0.1",
-      "resolved": "https://registry.npmjs.org/gopd/-/gopd-1.0.1.tgz",
-      "integrity": "sha512-d65bNlIadxvpb/A2abVdlqKqV563juRnZ1Wtk6s1sIR8uNsXR70xqIzVqxVf1eTqDunwT2MkczEeaezCKTZhwA==",
+      "version": "1.2.0",
+      "resolved": "https://registry.npmjs.org/gopd/-/gopd-1.2.0.tgz",
+      "integrity": "sha512-ZUKRh6/kUFoAiTAtTYPZJ3hw9wNxx+BIBOijnlG9PnrJsCcSjs1wyyD6vJpaYtgnzDrKYRSqf3OO6Rfa93xsRg==",
       "license": "MIT",
-      "dependencies": {
-        "get-intrinsic": "^1.1.3"
+      "engines": {
+        "node": ">= 0.4"
       },
       "funding": {
         "url": "https://github.com/sponsors/ljharb"
@@ -1261,22 +1592,10 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
-    "node_modules/has-proto": {
-      "version": "1.0.3",
-      "resolved": "https://registry.npmjs.org/has-proto/-/has-proto-1.0.3.tgz",
-      "integrity": "sha512-SJ1amZAJUiZS+PhsVLf5tGydlaVB8EdFpaSO4gmiUKUOxk8qzn5AIy4ZeJUmh22znIdk/uMAUT2pl3FxzVUH+Q==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 0.4"
-      },
-      "funding": {
-        "url": "https://github.com/sponsors/ljharb"
-      }
-    },
     "node_modules/has-symbols": {
-      "version": "1.0.3",
-      "resolved": "https://registry.npmjs.org/has-symbols/-/has-symbols-1.0.3.tgz",
-      "integrity": "sha512-l3LCuF6MgDNwTDKkdYGEihYjt5pRPbEg46rtlmnSPlUbgmB8LOIrKJbYYFBSbnPaJexMKtiPO8hmeRjRz2Td+A==",
+      "version": "1.1.0",
+      "resolved": "https://registry.npmjs.org/has-symbols/-/has-symbols-1.1.0.tgz",
+      "integrity": "sha512-1cDNdwJ2Jaohmb3sg4OmKaMBwuC48sYni5HUw2DvsC8LjGTLK9h+eb1X6RyuOHe4hT0ULCW68iomhjUoKUqlPQ==",
       "license": "MIT",
       "engines": {
         "node": ">= 0.4"
@@ -1327,6 +1646,16 @@
       "integrity": "sha512-8Sb3veuYCyrZL+VBt9LJfZjLUPWVvqn8tG28VqYNFCo43KHcKuq+b4EiXGeuaLAQWL2YmyDgMp2aSpH9JHsEQg==",
       "license": "Apache-2.0"
     },
+    "node_modules/ignore": {
+      "version": "5.3.2",
+      "resolved": "https://registry.npmjs.org/ignore/-/ignore-5.3.2.tgz",
+      "integrity": "sha512-hsBTNUqQTDwkWtcdYI2i06Y/nUBEsNEDJKjWdigLvegy8kDuJAS8uRlpkkcQpyEXL0Z/pjDy5HBmMjRCJ2gq+g==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">= 4"
+      }
+    },
     "node_modules/ignore-by-default": {
       "version": "1.0.1",
       "resolved": "https://registry.npmjs.org/ignore-by-default/-/ignore-by-default-1.0.1.tgz",
@@ -1431,6 +1760,18 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/is-interactive": {
+      "version": "2.0.0",
+      "resolved": "https://registry.npmjs.org/is-interactive/-/is-interactive-2.0.0.tgz",
+      "integrity": "sha512-qP1vozQRI+BMOPcjFzrjXuQvdak2pHNUMZoeG2eRbiSqyvbEf/wQtEOTOX1guk6E3t36RkaqiSt8A/6YElNxLQ==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=12"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/is-number": {
       "version": "7.0.0",
       "resolved": "https://registry.npmjs.org/is-number/-/is-number-7.0.0.tgz",
@@ -1456,6 +1797,18 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/is-unicode-supported": {
+      "version": "2.1.0",
+      "resolved": "https://registry.npmjs.org/is-unicode-supported/-/is-unicode-supported-2.1.0.tgz",
+      "integrity": "sha512-mE00Gnza5EEB3Ds0HfMyllZzbBrmLOX3vfWoj9A9PEnTfratQ/BcaJOuMhnkhjXvb2+FkY3VuHqtAGpTPmglFQ==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/is-url": {
       "version": "1.2.4",
       "resolved": "https://registry.npmjs.org/is-url/-/is-url-1.2.4.tgz",
@@ -1516,6 +1869,34 @@
         "node": ">=6"
       }
     },
+    "node_modules/log-symbols": {
+      "version": "6.0.0",
+      "resolved": "https://registry.npmjs.org/log-symbols/-/log-symbols-6.0.0.tgz",
+      "integrity": "sha512-i24m8rpwhmPIS4zscNzK6MSEhk0DUWa/8iYQWxhffV8jkI4Phvs3F+quL5xvS0gdQR0FyTCMMH33Y78dDTzzIw==",
+      "license": "MIT",
+      "dependencies": {
+        "chalk": "^5.3.0",
+        "is-unicode-supported": "^1.3.0"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
+    "node_modules/log-symbols/node_modules/is-unicode-supported": {
+      "version": "1.3.0",
+      "resolved": "https://registry.npmjs.org/is-unicode-supported/-/is-unicode-supported-1.3.0.tgz",
+      "integrity": "sha512-43r2mRvz+8JRIKnWJ+3j8JtjRKZ6GmjzfaE/qiBJnikNnYv/6bagRJ1kUhNk8R5EX/GkobD+r+sfxCPJsiKBLQ==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=12"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/lru-cache": {
       "version": "4.1.5",
       "resolved": "https://registry.npmjs.org/lru-cache/-/lru-cache-4.1.5.tgz",
@@ -1526,6 +1907,39 @@
         "yallist": "^2.1.2"
       }
     },
+    "node_modules/math-intrinsics": {
+      "version": "1.1.0",
+      "resolved": "https://registry.npmjs.org/math-intrinsics/-/math-intrinsics-1.1.0.tgz",
+      "integrity": "sha512-/IXtbwEk5HTPyEwyKX6hGkYXxM9nbj64B+ilVJnC/R6B0pH5G4V3b0pVbL7DBj4tkhBAppbQUlf6F6Xl9LHu1g==",
+      "license": "MIT",
+      "engines": {
+        "node": ">= 0.4"
+      }
+    },
+    "node_modules/merge2": {
+      "version": "1.4.1",
+      "resolved": "https://registry.npmjs.org/merge2/-/merge2-1.4.1.tgz",
+      "integrity": "sha512-8q7VEgMJW4J8tcfVPy8g09NcQwZdbwFEqhe/WZkoIzjn/3TGDwtOCYtXGxA3O8tPzpczCCDgv+P2P5y00ZJOOg==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">= 8"
+      }
+    },
+    "node_modules/micromatch": {
+      "version": "4.0.8",
+      "resolved": "https://registry.npmjs.org/micromatch/-/micromatch-4.0.8.tgz",
+      "integrity": "sha512-PXwfBhYu0hBCPw8Dn0E+WDYb7af3dSLVWKi3HGv84IdF4TyFoC0ysxFd0Goxw7nSv4T/PzEJQxsYsEiFCKo2BA==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "braces": "^3.0.3",
+        "picomatch": "^2.3.1"
+      },
+      "engines": {
+        "node": ">=8.6"
+      }
+    },
     "node_modules/mime-db": {
       "version": "1.52.0",
       "resolved": "https://registry.npmjs.org/mime-db/-/mime-db-1.52.0.tgz",
@@ -1547,12 +1961,38 @@
         "node": ">= 0.6"
       }
     },
+    "node_modules/mimic-function": {
+      "version": "5.0.1",
+      "resolved": "https://registry.npmjs.org/mimic-function/-/mimic-function-5.0.1.tgz",
+      "integrity": "sha512-VP79XUPxV2CigYP3jWwAUFSku2aKqBH7uTAapFWCBqutsbmDo96KY5o8uh6U+/YSIn5OxJnXp73beVkpqMIGhA==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/ms": {
       "version": "2.1.3",
       "resolved": "https://registry.npmjs.org/ms/-/ms-2.1.3.tgz",
       "integrity": "sha512-6FlzubTLZG3J2a/NVCAleEhjzq5oxgHyaCU9yYXvcLsvoVaHJq/s5xXI6/XXP6tz7R9xAOtHnSO/tXtF3WRTlA==",
       "license": "MIT"
     },
+    "node_modules/mylas": {
+      "version": "2.1.13",
+      "resolved": "https://registry.npmjs.org/mylas/-/mylas-2.1.13.tgz",
+      "integrity": "sha512-+MrqnJRtxdF+xngFfUUkIMQrUUL0KsxbADUkn23Z/4ibGg192Q+z+CQyiYwvWTsYjJygmMR8+w3ZDa98Zh6ESg==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">=12.0.0"
+      },
+      "funding": {
+        "type": "github",
+        "url": "https://github.com/sponsors/raouldeheer"
+      }
+    },
     "node_modules/nanoid": {
       "version": "3.3.8",
       "resolved": "https://registry.npmjs.org/nanoid/-/nanoid-3.3.8.tgz",
@@ -1691,6 +2131,21 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/onetime": {
+      "version": "7.0.0",
+      "resolved": "https://registry.npmjs.org/onetime/-/onetime-7.0.0.tgz",
+      "integrity": "sha512-VXJjc87FScF88uafS3JllDgvAm+c/Slfz06lorj2uAY34rlUu0Nt+v8wreiImcrgAjjIHp1rXpTDlLOGw29WwQ==",
+      "license": "MIT",
+      "dependencies": {
+        "mimic-function": "^5.0.0"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/openai": {
       "version": "4.72.0",
       "resolved": "https://registry.npmjs.org/openai/-/openai-4.72.0.tgz",
@@ -1741,6 +2196,29 @@
         "opencollective-postinstall": "index.js"
       }
     },
+    "node_modules/ora": {
+      "version": "8.2.0",
+      "resolved": "https://registry.npmjs.org/ora/-/ora-8.2.0.tgz",
+      "integrity": "sha512-weP+BZ8MVNnlCm8c0Qdc1WSWq4Qn7I+9CJGm7Qali6g44e/PUzbjNqJX5NJ9ljlNMosfJvg1fKEGILklK9cwnw==",
+      "license": "MIT",
+      "dependencies": {
+        "chalk": "^5.3.0",
+        "cli-cursor": "^5.0.0",
+        "cli-spinners": "^2.9.2",
+        "is-interactive": "^2.0.0",
+        "is-unicode-supported": "^2.0.0",
+        "log-symbols": "^6.0.0",
+        "stdin-discarder": "^0.2.2",
+        "string-width": "^7.2.0",
+        "strip-ansi": "^7.1.0"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/os": {
       "version": "0.1.2",
       "resolved": "https://registry.npmjs.org/os/-/os-0.1.2.tgz",
@@ -1778,6 +2256,16 @@
         "util": "^0.10.3"
       }
     },
+    "node_modules/path-type": {
+      "version": "4.0.0",
+      "resolved": "https://registry.npmjs.org/path-type/-/path-type-4.0.0.tgz",
+      "integrity": "sha512-gDKb8aZMDeD/tZWs9P6+q0J9Mwkdl6xMV8TjnGP3qJVJ06bdMgkbBlLU8IdfOsIsFz2BW1rNVT3XuNEl8zPAvw==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">=8"
+      }
+    },
     "node_modules/path/node_modules/inherits": {
       "version": "2.0.3",
       "resolved": "https://registry.npmjs.org/inherits/-/inherits-2.0.3.tgz",
@@ -1834,6 +2322,19 @@
         "url": "https://github.com/sponsors/jonschlinkert"
       }
     },
+    "node_modules/plimit-lit": {
+      "version": "1.6.1",
+      "resolved": "https://registry.npmjs.org/plimit-lit/-/plimit-lit-1.6.1.tgz",
+      "integrity": "sha512-B7+VDyb8Tl6oMJT9oSO2CW8XC/T4UcJGrwOVoNGwOQsQYhlpfajmrMj5xeejqaASq3V/EqThyOeATEOMuSEXiA==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "queue-lit": "^1.5.1"
+      },
+      "engines": {
+        "node": ">=12"
+      }
+    },
     "node_modules/possible-typed-array-names": {
       "version": "1.0.0",
       "resolved": "https://registry.npmjs.org/possible-typed-array-names/-/possible-typed-array-names-1.0.0.tgz",
@@ -1871,6 +2372,37 @@
       "dev": true,
       "license": "MIT"
     },
+    "node_modules/queue-lit": {
+      "version": "1.5.2",
+      "resolved": "https://registry.npmjs.org/queue-lit/-/queue-lit-1.5.2.tgz",
+      "integrity": "sha512-tLc36IOPeMAubu8BkW8YDBV+WyIgKlYU7zUNs0J5Vk9skSZ4JfGlPOqplP0aHdfv7HL0B2Pg6nwiq60Qc6M2Hw==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">=12"
+      }
+    },
+    "node_modules/queue-microtask": {
+      "version": "1.2.3",
+      "resolved": "https://registry.npmjs.org/queue-microtask/-/queue-microtask-1.2.3.tgz",
+      "integrity": "sha512-NuaNSa6flKT5JaSYQzJok04JzTL1CA6aGhv5rfLW3PgqA+M2ChpZQnAC8h8i4ZFkBS8X5RqkDBHA7r4hej3K9A==",
+      "dev": true,
+      "funding": [
+        {
+          "type": "github",
+          "url": "https://github.com/sponsors/feross"
+        },
+        {
+          "type": "patreon",
+          "url": "https://www.patreon.com/feross"
+        },
+        {
+          "type": "consulting",
+          "url": "https://feross.org/support"
+        }
+      ],
+      "license": "MIT"
+    },
     "node_modules/react": {
       "version": "19.0.0",
       "resolved": "https://registry.npmjs.org/react/-/react-19.0.0.tgz",
@@ -1900,6 +2432,67 @@
       "integrity": "sha512-kY1AZVr2Ra+t+piVaJ4gxaFaReZVH40AKNo7UCX6W+dEwBo/2oZJzqfuN1qLq1oL45o56cPaTXELwrTh8Fpggg==",
       "license": "MIT"
     },
+    "node_modules/resolve-pkg-maps": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/resolve-pkg-maps/-/resolve-pkg-maps-1.0.0.tgz",
+      "integrity": "sha512-seS2Tj26TBVOC2NIc2rOe2y2ZO7efxITtLZcGSOnHHNOQ7CkiUBfw0Iw2ck6xkIhPwLhKNLS8BO+hEpngQlqzw==",
+      "dev": true,
+      "license": "MIT",
+      "funding": {
+        "url": "https://github.com/privatenumber/resolve-pkg-maps?sponsor=1"
+      }
+    },
+    "node_modules/restore-cursor": {
+      "version": "5.1.0",
+      "resolved": "https://registry.npmjs.org/restore-cursor/-/restore-cursor-5.1.0.tgz",
+      "integrity": "sha512-oMA2dcrw6u0YfxJQXm342bFKX/E4sG9rbTzO9ptUcR/e8A33cHuvStiYOwH7fszkZlZ1z/ta9AAoPk2F4qIOHA==",
+      "license": "MIT",
+      "dependencies": {
+        "onetime": "^7.0.0",
+        "signal-exit": "^4.1.0"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
+    "node_modules/reusify": {
+      "version": "1.1.0",
+      "resolved": "https://registry.npmjs.org/reusify/-/reusify-1.1.0.tgz",
+      "integrity": "sha512-g6QUff04oZpHs0eG5p83rFLhHeV00ug/Yf9nZM6fLeUrPguBTkTQOdpAWWspMh55TZfVQDPaN3NQJfbVRAxdIw==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "iojs": ">=1.0.0",
+        "node": ">=0.10.0"
+      }
+    },
+    "node_modules/run-parallel": {
+      "version": "1.2.0",
+      "resolved": "https://registry.npmjs.org/run-parallel/-/run-parallel-1.2.0.tgz",
+      "integrity": "sha512-5l4VyZR86LZ/lDxZTR6jqL8AFE2S0IFLMP26AbjsLVADxHdhB/c0GUsH+y39UfCi3dzz8OlQuPmnaJOMoDHQBA==",
+      "dev": true,
+      "funding": [
+        {
+          "type": "github",
+          "url": "https://github.com/sponsors/feross"
+        },
+        {
+          "type": "patreon",
+          "url": "https://www.patreon.com/feross"
+        },
+        {
+          "type": "consulting",
+          "url": "https://feross.org/support"
+        }
+      ],
+      "license": "MIT",
+      "dependencies": {
+        "queue-microtask": "^1.2.2"
+      }
+    },
     "node_modules/secure-json-parse": {
       "version": "2.7.0",
       "resolved": "https://registry.npmjs.org/secure-json-parse/-/secure-json-parse-2.7.0.tgz",
@@ -1974,6 +2567,18 @@
         "@img/sharp-win32-x64": "0.33.5"
       }
     },
+    "node_modules/signal-exit": {
+      "version": "4.1.0",
+      "resolved": "https://registry.npmjs.org/signal-exit/-/signal-exit-4.1.0.tgz",
+      "integrity": "sha512-bzyZ1e88w9O1iNJbKnOlvYTrWPDl46O1bG0D3XInv+9tkPrxrN8jUUTiFlDkkmKWgn1M6CfIA13SuGqOa9Korw==",
+      "license": "ISC",
+      "engines": {
+        "node": ">=14"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/isaacs"
+      }
+    },
     "node_modules/simple-swizzle": {
       "version": "0.2.2",
       "resolved": "https://registry.npmjs.org/simple-swizzle/-/simple-swizzle-0.2.2.tgz",
@@ -1996,6 +2601,60 @@
         "node": ">=10"
       }
     },
+    "node_modules/slash": {
+      "version": "3.0.0",
+      "resolved": "https://registry.npmjs.org/slash/-/slash-3.0.0.tgz",
+      "integrity": "sha512-g9Q1haeby36OSStwb4ntCGGGaKsaVSjQ68fBxoQcutl5fS1vuY18H3wSt3jFyFtrkx+Kz0V1G85A4MyAdDMi2Q==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">=8"
+      }
+    },
+    "node_modules/stdin-discarder": {
+      "version": "0.2.2",
+      "resolved": "https://registry.npmjs.org/stdin-discarder/-/stdin-discarder-0.2.2.tgz",
+      "integrity": "sha512-UhDfHmA92YAlNnCfhmq0VeNL5bDbiZGg7sZ2IvPsXubGkiNa9EC+tUTsjBRsYUAz87btI6/1wf4XoVvQ3uRnmQ==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
+    "node_modules/string-width": {
+      "version": "7.2.0",
+      "resolved": "https://registry.npmjs.org/string-width/-/string-width-7.2.0.tgz",
+      "integrity": "sha512-tsaTIkKW9b4N+AEj+SVA+WhJzV7/zMhcSu78mLKWSk7cXMOSHsBKFWUs0fWwq8QyK3MgJBQRX6Gbi4kYbdvGkQ==",
+      "license": "MIT",
+      "dependencies": {
+        "emoji-regex": "^10.3.0",
+        "get-east-asian-width": "^1.0.0",
+        "strip-ansi": "^7.1.0"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
+    "node_modules/strip-ansi": {
+      "version": "7.1.2",
+      "resolved": "https://registry.npmjs.org/strip-ansi/-/strip-ansi-7.1.2.tgz",
+      "integrity": "sha512-gmBGslpoQJtgnMAvOVqGZpEz9dyoKTCzy2nfz/n8aIFhN/jCE/rCmcxabB6jOOHV+0WNnylOxaxBQPSvcWklhA==",
+      "license": "MIT",
+      "dependencies": {
+        "ansi-regex": "^6.0.1"
+      },
+      "engines": {
+        "node": ">=12"
+      },
+      "funding": {
+        "url": "https://github.com/chalk/strip-ansi?sponsor=1"
+      }
+    },
     "node_modules/supports-color": {
       "version": "5.5.0",
       "resolved": "https://registry.npmjs.org/supports-color/-/supports-color-5.5.0.tgz",
@@ -2097,6 +2756,38 @@
       "integrity": "sha512-N3WMsuqV66lT30CrXNbEjx4GEwlow3v6rr4mCcv6prnfwhS01rkgyFdjPNBYd9br7LpXV1+Emh01fHnq2Gdgrw==",
       "license": "MIT"
     },
+    "node_modules/tsc-alias": {
+      "version": "1.8.16",
+      "resolved": "https://registry.npmjs.org/tsc-alias/-/tsc-alias-1.8.16.tgz",
+      "integrity": "sha512-QjCyu55NFyRSBAl6+MTFwplpFcnm2Pq01rR/uxfqJoLMm6X3O14KEGtaSDZpJYaE1bJBGDjD0eSuiIWPe2T58g==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "chokidar": "^3.5.3",
+        "commander": "^9.0.0",
+        "get-tsconfig": "^4.10.0",
+        "globby": "^11.0.4",
+        "mylas": "^2.1.9",
+        "normalize-path": "^3.0.0",
+        "plimit-lit": "^1.2.6"
+      },
+      "bin": {
+        "tsc-alias": "dist/bin/index.js"
+      },
+      "engines": {
+        "node": ">=16.20.2"
+      }
+    },
+    "node_modules/tsc-alias/node_modules/commander": {
+      "version": "9.5.0",
+      "resolved": "https://registry.npmjs.org/commander/-/commander-9.5.0.tgz",
+      "integrity": "sha512-KRs7WVDKg86PWiuAqhDrAQnTXZKraVcCc6vFdL14qrZ/DcWwuRo7VoiYXalXO7S5GKpqYiVEwCbgFDfxNHKJBQ==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": "^12.20.0 || >=14"
+      }
+    },
     "node_modules/tslib": {
       "version": "1.14.1",
       "resolved": "https://registry.npmjs.org/tslib/-/tslib-1.14.1.tgz",
diff --git a/package.json b/package.json
index 336c9cd..155a499 100644
--- a/package.json
+++ b/package.json
@@ -3,14 +3,22 @@
   "version": "1.1.5",
   "description": "Intelligent document processing and extraction.",
   "scripts": {
-    "start": "npm run start --workspace=extractor",
-    "dev": "npm run dev --workspace=extractor"
+    "build": "npm run build:core && npm run build:extractor && npm run build:cli",
+    "build:core": "npm run build --workspace=core",
+    "build:extractor": "npm run build --workspace=extractor",
+    "build:cli": "npm run build --workspace=cli",
+    "prepublishOnly": "npm run build",
+    "start": "npm run start --workspace=cli",
+    "dev": "npm run dev --workspace=cli",
+    "cli": "npm run start --workspace=cli"
   },
   "workspaces": [
     "core",
-    "extractor"
+    "extractor",
+    "cli"
   ],
-  "main": "./extractor/src/index.js",
+  "main": "./extractor/dist/index.js",
+  "types": "./extractor/dist/index.d.ts",
   "keywords": [
     "document",
     "extraction",
@@ -25,7 +33,7 @@
   },
   "homepage": "https://documind.xyz",
   "dependencies": {
-    "axios": "^1.7.7",
+    "axios": "^1.12.2",
     "child_process": "^1.0.2",
     "core": "file:./core",
     "dotenv": "^16.4.5",