{ "cells": [ { "cell_type": "markdown", "id": "0", "metadata": { "lines_to_next_cell": 0 }, "source": [ "# File Converters\n", "\n", "File converters transform text into file outputs such as PDFs. These converters are useful for packaging prompts into distributable formats.\n", "\n", "## Overview\n", "\n", "This notebook covers:\n", "\n", "- **PDFConverter**: Convert text to PDF documents with templates or direct generation\n", "- **WordDocConverter**: Convert text to Word (.docx) documents with optional placeholder injection" ] }, { "cell_type": "markdown", "id": "1", "metadata": { "lines_to_next_cell": 0 }, "source": [ "## PDFConverter\n", "\n", "The `PDFConverter` generates PDF documents from text in multiple modes:\n", "\n", "1. **Template-Based PDF Generation**: Use YAML templates to render dynamic content into PDFs\n", "2. **Direct Prompt PDF Generation**: Convert plain text strings into PDFs without templates\n", "3. **Modify Existing PDFs**: Inject text into existing PDF documents" ] }, { "cell_type": "markdown", "id": "2", "metadata": { "lines_to_next_cell": 0 }, "source": [ "### Template-Based PDF Generation\n", "\n", "This mode populates placeholders in a YAML-based template and converts the rendered content into a PDF." ] }, { "cell_type": "code", "execution_count": null, "id": "3", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Found default environment files: ['./.pyrit/.env', './.pyrit/.env.local']\n", "Loaded environment file: ./.pyrit/.env\n", "Loaded environment file: ./.pyrit/.env.local\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "No new upgrade operations detected.\n", "TextTarget: user: ./dbdata/prompt-memory-entries/binaries/1779402999644083.pdf\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\u001b[1m\u001b[33m ❓ ATTACK RESULT: UNDETERMINED ❓ \u001b[0m\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Attack Summary \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m 📋 Basic Information\u001b[0m\n", "\u001b[36m • Objective: {'hiring_manager_name': 'Jane Doe', 'current_role': 'AI Engineer', 'company': 'CyberDefense Inc.', 'red_teaming_reason': 'to creatively identify security vulnerabilities while enjoying free coffee', 'applicant_name': 'John Smith'}\u001b[0m\n", "\u001b[36m • Attack Type: PromptSendingAttack\u001b[0m\n", "\u001b[36m • Conversation ID: 9edbd5d0-2bde-4432-b4a7-91b9688f4377\u001b[0m\n", "\n", "\u001b[1m ⚡ Execution Metrics\u001b[0m\n", "\u001b[32m • Turns Executed: 1\u001b[0m\n", "\u001b[32m • Execution Time: 45ms\u001b[0m\n", "\n", "\u001b[1m 🎯 Outcome\u001b[0m\n", "\u001b[33m • Status: ❓ UNDETERMINED\u001b[0m\n", "\u001b[37m • Reason: No objective scorer configured\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Conversation History with Objective Target \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m\u001b[34m🔹 Turn 1 - USER\u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[36m Original:\u001b[0m\n", "\u001b[37m {'hiring_manager_name': 'Jane Doe', 'current_role': 'AI Engineer', 'company': 'CyberDefense Inc.',\u001b[0m\n", "\u001b[37m 'red_teaming_reason': 'to creatively identify security vulnerabilities while enjoying free\u001b[0m\n", "\u001b[37m coffee', 'applicant_name': 'John Smith'}\u001b[0m\n", "\n", "\u001b[36m Converted:\u001b[0m\n", "\u001b[37m ./dbdata/prompt-\u001b[0m\n", "\u001b[37m memory-entries\\binaries\\1779402999644083.pdf\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[2m\u001b[37m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[2m\u001b[37m Report generated at: 2026-05-21 22:36:39 UTC \u001b[0m\n" ] } ], "source": [ "import pathlib\n", "\n", "from pyrit.common.path import CONVERTER_SEED_PROMPT_PATH\n", "from pyrit.converter import PDFConverter\n", "from pyrit.executor.attack import (\n", " AttackConverterConfig,\n", " PromptSendingAttack,\n", ")\n", "from pyrit.models import SeedPrompt\n", "from pyrit.output import output_attack_async\n", "from pyrit.prompt_normalizer import ConverterConfiguration\n", "from pyrit.prompt_target import TextTarget\n", "from pyrit.setup import IN_MEMORY, initialize_pyrit_async\n", "\n", "await initialize_pyrit_async(memory_db_type=IN_MEMORY) # type: ignore\n", "\n", "prompt_data = {\n", " \"hiring_manager_name\": \"Jane Doe\",\n", " \"current_role\": \"AI Engineer\",\n", " \"company\": \"CyberDefense Inc.\",\n", " \"red_teaming_reason\": \"to creatively identify security vulnerabilities while enjoying free coffee\",\n", " \"applicant_name\": \"John Smith\",\n", "}\n", "\n", "# Load the YAML template for the PDF generation\n", "template_path = pathlib.Path(CONVERTER_SEED_PROMPT_PATH) / \"pdf_converters\" / \"red_teaming_application_template.yaml\"\n", "if not template_path.exists():\n", " raise FileNotFoundError(f\"Template file not found: {template_path}\")\n", "\n", "# Load the SeedPrompt from the YAML file\n", "prompt_template = SeedPrompt.from_yaml_file(template_path)\n", "\n", "# Initialize target\n", "prompt_target = TextTarget()\n", "\n", "# Initialize the PDFConverter\n", "pdf_converter = ConverterConfiguration.from_converters(\n", " converters=[\n", " PDFConverter(\n", " prompt_template=prompt_template,\n", " font_type=\"Arial\",\n", " font_size=12,\n", " page_width=210,\n", " page_height=297,\n", " )\n", " ]\n", ")\n", "\n", "converter_config = AttackConverterConfig(\n", " request_converters=pdf_converter,\n", ")\n", "\n", "# Define prompt for the attack\n", "prompt = str(prompt_data)\n", "\n", "# Initialize the attack\n", "attack = PromptSendingAttack(\n", " objective_target=prompt_target,\n", " attack_converter_config=converter_config,\n", ")\n", "\n", "result = await attack.execute_async(objective=prompt) # type: ignore\n", "await output_attack_async(result)" ] }, { "cell_type": "markdown", "id": "4", "metadata": {}, "source": [ "### Direct Prompt PDF Generation (No Template)\n", "\n", "This mode converts plain text strings directly into PDFs without using templates." ] }, { "cell_type": "code", "execution_count": null, "id": "5", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "TextTarget: user: ./dbdata/prompt-memory-entries/binaries/1779402999698469.pdf\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\u001b[1m\u001b[33m ❓ ATTACK RESULT: UNDETERMINED ❓ \u001b[0m\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Attack Summary \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m 📋 Basic Information\u001b[0m\n", "\u001b[36m • Objective: This is a simple test string for PDF generation. No templates here!\u001b[0m\n", "\u001b[36m • Attack Type: PromptSendingAttack\u001b[0m\n", "\u001b[36m • Conversation ID: 3d86d651-30f2-4d2d-8585-a130ba3dc077\u001b[0m\n", "\n", "\u001b[1m ⚡ Execution Metrics\u001b[0m\n", "\u001b[32m • Turns Executed: 1\u001b[0m\n", "\u001b[32m • Execution Time: 23ms\u001b[0m\n", "\n", "\u001b[1m 🎯 Outcome\u001b[0m\n", "\u001b[33m • Status: ❓ UNDETERMINED\u001b[0m\n", "\u001b[37m • Reason: No objective scorer configured\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Conversation History with Objective Target \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m\u001b[34m🔹 Turn 1 - USER\u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[36m Original:\u001b[0m\n", "\u001b[37m This is a simple test string for PDF generation. No templates here!\u001b[0m\n", "\n", "\u001b[36m Converted:\u001b[0m\n", "\u001b[37m ./dbdata/prompt-\u001b[0m\n", "\u001b[37m memory-entries\\binaries\\1779402999698469.pdf\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[2m\u001b[37m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[2m\u001b[37m Report generated at: 2026-05-21 22:36:39 UTC \u001b[0m\n" ] } ], "source": [ "# Define a simple string prompt (no templates)\n", "prompt = \"This is a simple test string for PDF generation. No templates here!\"\n", "\n", "# Initialize the PDFConverter without a template\n", "pdf_converter = ConverterConfiguration.from_converters(\n", " converters=[\n", " PDFConverter(\n", " prompt_template=None, # No template provided\n", " font_type=\"Arial\",\n", " font_size=12,\n", " page_width=210,\n", " page_height=297,\n", " )\n", " ]\n", ")\n", "\n", "converter_config = AttackConverterConfig(\n", " request_converters=pdf_converter,\n", ")\n", "\n", "# Initialize the attack\n", "attack = PromptSendingAttack(\n", " objective_target=prompt_target,\n", " attack_converter_config=converter_config,\n", ")\n", "\n", "result = await attack.execute_async(objective=prompt) # type: ignore\n", "await output_attack_async(result)" ] }, { "cell_type": "markdown", "id": "6", "metadata": {}, "source": [ "### Modifying Existing PDFs with Injection Items\n", "\n", "The `PDFConverter` can also inject text into existing PDF documents at specified locations." ] }, { "cell_type": "code", "execution_count": null, "id": "7", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[15:36:40][10][ai-red-team][INFO][Processing page 0 with 2 injection items.]\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "[15:36:40][13][ai-red-team][INFO][Processing page 1 with 2 injection items.]\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "[15:36:40][15][ai-red-team][INFO][Processing page 2 with 2 injection items.]\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "TextTarget: user: ./dbdata/prompt-memory-entries/binaries/1779403000017242.pdf\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\u001b[1m\u001b[33m ❓ ATTACK RESULT: UNDETERMINED ❓ \u001b[0m\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Attack Summary \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m 📋 Basic Information\u001b[0m\n", "\u001b[36m • Objective: Modify existing PDF\u001b[0m\n", "\u001b[36m • Attack Type: PromptSendingAttack\u001b[0m\n", "\u001b[36m • Conversation ID: 7f9101d9-93b6-49e3-88ae-1bebaf44a0af\u001b[0m\n", "\n", "\u001b[1m ⚡ Execution Metrics\u001b[0m\n", "\u001b[32m • Turns Executed: 1\u001b[0m\n", "\u001b[32m • Execution Time: 31ms\u001b[0m\n", "\n", "\u001b[1m 🎯 Outcome\u001b[0m\n", "\u001b[33m • Status: ❓ UNDETERMINED\u001b[0m\n", "\u001b[37m • Reason: No objective scorer configured\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Conversation History with Objective Target \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m\u001b[34m🔹 Turn 1 - USER\u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[36m Original:\u001b[0m\n", "\u001b[37m Modify existing PDF\u001b[0m\n", "\n", "\u001b[36m Converted:\u001b[0m\n", "\u001b[37m ./dbdata/prompt-\u001b[0m\n", "\u001b[37m memory-entries\\binaries\\1779403000017242.pdf\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[2m\u001b[37m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[2m\u001b[37m Report generated at: 2026-05-21 22:36:40 UTC \u001b[0m\n" ] } ], "source": [ "import tempfile\n", "from pathlib import Path\n", "\n", "import requests\n", "\n", "# Download a sample PDF\n", "url = \"https://raw.githubusercontent.com/microsoft/PyRIT/main/pyrit/datasets/converters/pdf_converters/fake_CV.pdf\"\n", "\n", "with tempfile.NamedTemporaryFile(delete=False, suffix=\".pdf\") as tmp_file:\n", " response = requests.get(url)\n", " tmp_file.write(response.content)\n", "\n", "cv_pdf_path = Path(tmp_file.name)\n", "\n", "# Define injection items\n", "injection_items = [\n", " {\n", " \"page\": 0,\n", " \"x\": 50,\n", " \"y\": 700,\n", " \"text\": \"Injected Text\",\n", " \"font_size\": 12,\n", " \"font\": \"Helvetica\",\n", " \"font_color\": (255, 0, 0),\n", " }, # Red text\n", " {\n", " \"page\": 1,\n", " \"x\": 100,\n", " \"y\": 600,\n", " \"text\": \"Confidential\",\n", " \"font_size\": 10,\n", " \"font\": \"Helvetica\",\n", " \"font_color\": (0, 0, 255),\n", " }, # Blue text\n", "]\n", "\n", "# Initialize the PDFConverter with the existing PDF and injection items\n", "pdf_converter = ConverterConfiguration.from_converters(\n", " converters=[\n", " PDFConverter(\n", " prompt_template=None,\n", " font_type=\"Arial\",\n", " font_size=12,\n", " page_width=210,\n", " page_height=297,\n", " existing_pdf=cv_pdf_path, # Provide the existing PDF\n", " injection_items=injection_items, # Provide the injection items\n", " )\n", " ]\n", ")\n", "\n", "converter_config = AttackConverterConfig(\n", " request_converters=pdf_converter,\n", ")\n", "\n", "# Initialize the attack\n", "attack = PromptSendingAttack(\n", " objective_target=prompt_target,\n", " attack_converter_config=converter_config,\n", ")\n", "\n", "result = await attack.execute_async(objective=\"Modify existing PDF\") # type: ignore\n", "await output_attack_async(result)" ] }, { "cell_type": "markdown", "id": "8", "metadata": {}, "source": [ "## WordDocConverter\n", "\n", "The `WordDocConverter` generates Word (.docx) documents from text. It supports two main modes:\n", "\n", "1. **New Document Generation**: Convert plain text strings into Word documents\n", "2. **Placeholder-Based Injection**: Replace a placeholder in an existing `.docx` document with rendered content\n", "\n", "Placeholders must be fully contained within a single run in the document. If a placeholder spans\n", "multiple runs (e.g., due to mixed formatting), it will not be replaced." ] }, { "cell_type": "markdown", "id": "9", "metadata": {}, "source": [ "### Direct Word Document Generation (No Template)\n", "\n", "This mode converts plain text strings directly into `.docx` files." ] }, { "cell_type": "code", "execution_count": null, "id": "10", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "TextTarget: user: ./dbdata/prompt-memory-entries/binaries/1779403000064406.docx\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\u001b[1m\u001b[33m ❓ ATTACK RESULT: UNDETERMINED ❓ \u001b[0m\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Attack Summary \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m 📋 Basic Information\u001b[0m\n", "\u001b[36m • Objective: This is a simple test string for Word document generation.\u001b[0m\n", "\u001b[36m • Attack Type: PromptSendingAttack\u001b[0m\n", "\u001b[36m • Conversation ID: 172b3107-93fa-4443-b155-3d94f41e79ea\u001b[0m\n", "\n", "\u001b[1m ⚡ Execution Metrics\u001b[0m\n", "\u001b[32m • Turns Executed: 1\u001b[0m\n", "\u001b[32m • Execution Time: 38ms\u001b[0m\n", "\n", "\u001b[1m 🎯 Outcome\u001b[0m\n", "\u001b[33m • Status: ❓ UNDETERMINED\u001b[0m\n", "\u001b[37m • Reason: No objective scorer configured\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Conversation History with Objective Target \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m\u001b[34m🔹 Turn 1 - USER\u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[36m Original:\u001b[0m\n", "\u001b[37m This is a simple test string for Word document generation.\u001b[0m\n", "\n", "\u001b[36m Converted:\u001b[0m\n", "\u001b[37m ./dbdata/prompt-\u001b[0m\n", "\u001b[37m memory-entries\\binaries\\1779403000064406.docx\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[2m\u001b[37m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[2m\u001b[37m Report generated at: 2026-05-21 22:36:40 UTC \u001b[0m\n" ] } ], "source": [ "from pyrit.converter import WordDocConverter\n", "\n", "word_doc_converter = ConverterConfiguration.from_converters(converters=[WordDocConverter()])\n", "\n", "converter_config = AttackConverterConfig(\n", " request_converters=word_doc_converter,\n", ")\n", "\n", "attack = PromptSendingAttack(\n", " objective_target=prompt_target,\n", " attack_converter_config=converter_config,\n", ")\n", "\n", "result = await attack.execute_async(objective=\"This is a simple test string for Word document generation.\") # type: ignore\n", "await output_attack_async(result)" ] }, { "cell_type": "markdown", "id": "11", "metadata": {}, "source": [ "### Placeholder-Based Injection into Existing Word Documents\n", "\n", "This mode replaces a literal placeholder string in an existing `.docx` file with the prompt content." ] }, { "cell_type": "code", "execution_count": null, "id": "12", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "TextTarget: user: ./dbdata/prompt-memory-entries/binaries/1779403000154543.docx\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\u001b[1m\u001b[33m ❓ ATTACK RESULT: UNDETERMINED ❓ \u001b[0m\n", "\u001b[33m════════════════════════════════════════════════════════════════════════════════════════════════════\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Attack Summary \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m 📋 Basic Information\u001b[0m\n", "\u001b[36m • Objective: AI Red Team Engineer\u001b[0m\n", "\u001b[36m • Attack Type: PromptSendingAttack\u001b[0m\n", "\u001b[36m • Conversation ID: f68f05e7-bb47-4a9d-93b1-aae35b0ba497\u001b[0m\n", "\n", "\u001b[1m ⚡ Execution Metrics\u001b[0m\n", "\u001b[32m • Turns Executed: 1\u001b[0m\n", "\u001b[32m • Execution Time: 54ms\u001b[0m\n", "\n", "\u001b[1m 🎯 Outcome\u001b[0m\n", "\u001b[33m • Status: ❓ UNDETERMINED\u001b[0m\n", "\u001b[37m • Reason: No objective scorer configured\u001b[0m\n", "\n", "\u001b[1m\u001b[44m\u001b[37m Conversation History with Objective Target \u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[1m\u001b[34m🔹 Turn 1 - USER\u001b[0m\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[36m Original:\u001b[0m\n", "\u001b[37m AI Red Team Engineer\u001b[0m\n", "\n", "\u001b[36m Converted:\u001b[0m\n", "\u001b[37m ./dbdata/prompt-\u001b[0m\n", "\u001b[37m memory-entries\\binaries\\1779403000154543.docx\u001b[0m\n", "\n", "\u001b[34m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\n", "\u001b[2m\u001b[37m────────────────────────────────────────────────────────────────────────────────────────────────────\u001b[0m\n", "\u001b[2m\u001b[37m Report generated at: 2026-05-21 22:36:40 UTC \u001b[0m\n" ] } ], "source": [ "import tempfile\n", "\n", "from docx import Document # type: ignore[import-untyped]\n", "\n", "# Create a sample Word document with a placeholder\n", "with tempfile.NamedTemporaryFile(delete=False, suffix=\".docx\") as tmp_file:\n", " doc = Document()\n", " doc.add_paragraph(\"Dear Hiring Manager,\")\n", " doc.add_paragraph(\"I am writing to apply for the {{INJECTION_PLACEHOLDER}} position.\")\n", " doc.add_paragraph(\"Sincerely, Applicant\")\n", " doc.save(tmp_file.name)\n", " template_docx_path = Path(tmp_file.name)\n", "\n", "word_doc_converter = ConverterConfiguration.from_converters(\n", " converters=[\n", " WordDocConverter(\n", " existing_docx=template_docx_path,\n", " placeholder=\"{{INJECTION_PLACEHOLDER}}\",\n", " )\n", " ]\n", ")\n", "\n", "converter_config = AttackConverterConfig(\n", " request_converters=word_doc_converter,\n", ")\n", "\n", "attack = PromptSendingAttack(\n", " objective_target=prompt_target,\n", " attack_converter_config=converter_config,\n", ")\n", "\n", "result = await attack.execute_async(objective=\"AI Red Team Engineer\") # type: ignore\n", "await output_attack_async(result)" ] } ], "metadata": { "jupytext": { "cell_metadata_filter": "-all" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.11.9" } }, "nbformat": 4, "nbformat_minor": 5 }