import logging import os import traceback import pandas as pd import streamlit as st from llm_guard.vault import Vault from output import init_settings as init_output_settings from output import scan as scan_output from prompt import init_settings as init_prompt_settings from prompt import scan as scan_prompt PROMPT = "prompt" OUTPUT = "output" vault = Vault() st.set_page_config( page_title="LLM Guard Playground", layout="wide", initial_sidebar_state="expanded", menu_items={ "About": "https://llm-guard.com/", }, ) logger = logging.getLogger("llm-guard-playground") logger.setLevel(logging.INFO) # Sidebar st.sidebar.header( """ Scanning prompt and output using [LLM Guard](https://llm-guard.com/) """ ) scanner_type = st.sidebar.selectbox("Type", [PROMPT, OUTPUT], index=0) st_fail_fast = st.sidebar.checkbox( "Fail fast", value=False, help="Stop scanning after first failure" ) enabled_scanners = None settings = None if scanner_type == PROMPT: enabled_scanners, settings = init_prompt_settings() elif scanner_type == OUTPUT: enabled_scanners, settings = init_output_settings() # Main pannel st.subheader("Guard Prompt" if scanner_type == PROMPT else "Guard Output") with st.expander("About", expanded=False): st.info( """LLM-Guard is a comprehensive tool designed to fortify the security of Large Language Models (LLMs). \n\n[Code](https://github.com/protectai/llm-guard) | [Documentation](https://llm-guard.com/)""" ) analyzer_load_state = st.info("Starting LLM Guard...") analyzer_load_state.empty() # Before: prompt_examples_folder = "./examples/prompt" output_examples_folder = "./examples/output" prompt_examples = [f for f in os.listdir(prompt_examples_folder) if f.endswith(".txt")] output_examples = [f for f in os.listdir(output_examples_folder) if f.endswith(".txt")] if scanner_type == PROMPT: st_prompt_example = st.selectbox("Select prompt example", prompt_examples, index=0) with open(os.path.join(prompt_examples_folder, st_prompt_example), "r") as file: prompt_example_text = file.read() st_prompt_text = st.text_area( label="Enter prompt", value=prompt_example_text, height=200, key="prompt_text_input" ) elif scanner_type == OUTPUT: col1, col2 = st.columns(2) st_prompt_example = col1.selectbox("Select prompt example", prompt_examples, index=0) with open(os.path.join(prompt_examples_folder, st_prompt_example), "r") as file: prompt_example_text = file.read() st_prompt_text = col1.text_area( label="Enter prompt", value=prompt_example_text, height=300, key="prompt_text_input" ) st_output_example = col2.selectbox("Select output example", output_examples, index=0) with open(os.path.join(output_examples_folder, st_output_example), "r") as file: output_example_text = file.read() st_output_text = col2.text_area( label="Enter output", value=output_example_text, height=300, key="output_text_input" ) st_result_text = None st_analysis = None st_is_valid = None try: with st.form("text_form", clear_on_submit=False): submitted = st.form_submit_button("Process") if submitted: results = {} if scanner_type == PROMPT: st_result_text, results = scan_prompt( vault, enabled_scanners, settings, st_prompt_text, st_fail_fast ) elif scanner_type == OUTPUT: st_result_text, results = scan_output( vault, enabled_scanners, settings, st_prompt_text, st_output_text, st_fail_fast ) st_is_valid = all(item["is_valid"] for item in results) st_analysis = results except Exception as e: logger.error(e) traceback.print_exc() st.error(e) # After: if st_is_valid is not None: st.subheader(f"Results - {'valid' if st_is_valid else 'invalid'}") col1, col2 = st.columns(2) with col1: st.text_area(label="Sanitized text", value=st_result_text, height=400) with col2: st.table(pd.DataFrame(st_analysis))