pdf-ocr-cli/package.json

A CLI tool for OCR processing of PDF files using Mistral API with optional LLM verification

{
  "name": "pdf-ocr-cli",
  "version": "1.0.1",
  "description": "A CLI tool for OCR processing of PDF files using Mistral API with optional LLM verification",
  "main": "dist/cli.js",
  "bin": {
    "pdf-ocr": "./dist/cli.js"
  },
  "scripts": {
    "test": "jest",
    "test:watch": "jest --watch",
    "test:sample": "node test/scripts/test-sample.js",
    "test:multi-page": "node test/scripts/test-multi-page.js",
    "test:verify": "node test/scripts/verify-multi-page.js",
    "test:all": "npm run test && npm run test:sample && npm run test:multi-page && npm run test:verify",
    "build": "tsc",
    "start": "node dist/cli.js",
    "dev": "ts-node src/cli.ts",
    "prepublishOnly": "npm run build"
  },
  "keywords": [
    "pdf",
    "ocr",
    "cli",
    "mistral",
    "ai",
    "text-extraction",
    "pdf-processing",
    "togetherai",
    "llm"
  ],
  "author": "Luandro <luandro@gmail.com>",
  "license": "ISC",
  "repository": {
    "type": "git",
    "url": "git+https://github.com/luandro/pdf-ocr.git"
  },
  "bugs": {
    "url": "https://github.com/luandro/pdf-ocr/issues"
  },
  "homepage": "https://github.com/luandro/pdf-ocr#readme",
  "dependencies": {
    "@mistralai/mistralai": "^1.6.0",
    "@web-std/file": "^3.0.3",
    "@web-std/form-data": "^3.1.0",
    "canvas": "^3.1.0",
    "commander": "^13.1.0",
    "dotenv": "^16.5.0",
    "form-data": "^4.0.2",
    "formdata-node": "^6.0.3",
    "node-fetch": "^2.7.0",
    "pdf-lib": "^1.17.1",
    "pdf2pic": "^3.1.4",
    "sharp": "^0.34.1",
    "together-ai": "^0.16.0",
    "uuid": "^11.1.0"
  },
  "optionalDependencies": {
  },
  "devDependencies": {
    "@types/form-data": "^2.2.1",
    "@types/jest": "^29.5.14",
    "@types/node": "^22.15.3",
    "@types/node-fetch": "^2.6.12",
    "@types/uuid": "^10.0.0",
    "jest": "^29.7.0",
    "ts-jest": "^29.3.2",
    "ts-node": "^10.9.2",
    "typescript": "^5.8.3"
  }
}