first commit

.gitignore

@@ -0,0 +1 @@
+venv/*

README.md

@@ -0,0 +1,52 @@
+# LLM Security Chatbot
+
+The LLM Security Chatbot is a cutting-edge tool designed to assist in understanding and researching cybersecurity vulnerabilities. Built with the powerful Llama Large Language Model (LLM) and integrated into a user-friendly interface using Streamlit, this chatbot leverages natural language processing to provide in-depth analysis, actionable insights, and potential mitigation strategies for a wide range of security concerns.
+
+## Features
+
+- **Interactive Chat Interface**: Engage in conversational queries and receive detailed responses.
+- **Code Snippet Support**: Get examples and explanations with formatted code (still a work in progress) snippets for technical understanding.
+- **Conversation History**: Review past queries and responses directly within the application.
+- **Exportable Conversations**: Easily export the conversation history for documentation or further analysis.
+
+## Getting Started
+
+To get started with the LLM Security Chatbot, follow these steps:
+
+### Prerequisites
+
+Ensure you have Python 3.9+ installed on your system. You will also need the `streamlit` and `llama_cpp` packages.
+
+### Installation
+
+1. Clone the repository:
+    ```bash
+    git clone https://github.com/jwalker/llm_security_chatbot.git
+    ```
+2. Navigate to the project directory:
+    ```bash
+    cd llm_security_chatbot/
+    ```
+3. Install the required Python packages:
+    ```bash
+    uv pip install -r requirements.txt
+    ```
+
+### Running the Application
+
+Launch the chatbot by running the Streamlit application:
+
+```bash
+streamlit run app.py
+```
+
+Visit http://localhost:8501 in your web browser to start interacting with the chatbot.
+
+### Usage
+Enter your cybersecurity-related queries in the text area and hit 'Submit' to receive a response. The chat interface allows for natural language questions and provides detailed explanations, including code examples when relevant.
+
+### Contributing
+Contributions are welcome! If you have suggestions for improvements or want to contribute to the development of the LLM Security Chatbot, please feel free to fork the repository and submit a pull request.
+
+### Contact
+If you have any questions or feedback, please reach out via email: [email protected]

app.py

@@ -0,0 +1,124 @@
+import streamlit as st
+from llama_cpp import Llama
+
+# Initialize the Llama model
+llm = Llama(
+    model_path="../llama.cpp/models/mistral-7B-v0.1/ggml-model-Q4_K_M.gguf",
+    n_ctx=4096,
+    n_gpu_layers=-1,
+    chat_format="chatml"
+)
+
+st.set_page_config(page_title="💬 Security ChatBot")
+st.title('Security ChatBot - Ask me something')
+
+if 'conversation_history' not in st.session_state:
+    st.session_state['conversation_history'] = []
+if 'debug_info' not in st.session_state:
+    st.session_state['debug_info'] = {}
+
+user_query = st.text_area("Enter your query related to cybersecurity research", '')
+
+prompt_template = """
+Q: {user_query}
+Context: The user is seeking in-depth analysis and actionable insights on cybersecurity vulnerabilities. 
+Please provide detailed information, potential mitigation strategies, and reference relevant tools or resources.
+A: """
+
+def display_conversation():
+    """
+    Display the conversation history with the most recent messages first.
+    Format messages appropriately, including code blocks.
+    """
+    for exchange in st.session_state['conversation_history']:
+        # Check for the type of exchange to handle string-only history entries gracefully
+        if isinstance(exchange, dict):
+            message, sender = exchange['message'], exchange['sender']
+        else:  # Fallback for string entries, assuming all older entries are from the user
+            message, sender = exchange, 'User'
+
+        # Format message as Markdown to properly display code blocks and maintain line breaks
+        if sender == 'Agent' and '```' in message:
+            # Assuming code blocks are enclosed in triple backticks
+            st.markdown(f"**{sender}:**\n```{message.split('```')[1]}```", unsafe_allow_html=True)
+        else:
+            st.markdown(f"**{sender}:**\n{message}", unsafe_allow_html=True) # clearly not safe
+
+def display_debug_info():
+    if st.session_state['debug_info']:
+        st.json(st.session_state['debug_info'])
+
+def export_conversation_history():
+    with open('conversation_history.txt', 'w') as file:
+        for exchange in reversed(st.session_state['conversation_history']):
+            if isinstance(exchange, dict):
+                line = f"{exchange['sender']}: {exchange['message']}\n"
+            else:
+                line = f"User: {exchange}\n"
+            file.write(line)
+    st.success('Conversation exported successfully!')
+
+if st.button('Submit'):
+    if user_query:
+        st.session_state['conversation_history'].append({"sender": "Researcher", "message": user_query})
+        with st.spinner('Analyzing your query...'):
+            prompt = prompt_template.format(user_query=user_query)
+            output = llm(prompt, max_tokens=2048, stop=["Q:"], echo=True)
+
+            if 'choices' in output and len(output['choices']) > 0:
+                raw_response = output['choices'][0]['text']
+                user_friendly_response = raw_response.split('A: ')[-1].strip()
+
+                st.session_state['conversation_history'].append({"sender": "Agent", "message": user_friendly_response})
+                st.session_state['debug_info'] = output
+            else:
+                st.error("The model did not return a valid response. Please try again.")
+    else:
+        st.warning("Please enter a query.")
+
+display_conversation()
+# Sidebar for additional controls
+with st.sidebar:
+    st.image('https://files.oaiusercontent.com/file-TAKOjgzaq5efA7OIusxh2Vkw?se=2024-03-20T06%3A55%3A23Z&sp=r&sv=2021-08-06&sr=b&rscc=max-age%3D31536000%2C%20immutable&rscd=attachment%3B%20filename%3Dbf88bd85-a472-465b-8707-3d315307bc9b.webp&sig=CuqICRvq4pHQ45NGIxADC5AcIjrzdutbcrrYLKA73AY%3D', width=100)  # Consider adding a logo or related visual
+    st.markdown('## 🛠 Controls & Tools')
+    st.markdown("""---""") 
+
+    # Reset Conversation Button
+    col1, col2 = st.columns([1, 4])
+    with col1:
+        if st.button('🔄', help='Reset Conversation'):
+            st.session_state.conversation_history = []
+            st.session_state.debug_info = {}
+    with col2:
+        st.markdown("**Reset Conversation**")
+
+    # Export Conversation History Button
+    col1, col2 = st.columns([1, 4])
+    with col1:
+        if st.button('💾', help='Export Conversation History'):
+            export_conversation_history()
+    with col2:
+        st.markdown("**Export Conversation History**")
+
+    # Show Debug Information Button
+    col1, col2 = st.columns([1, 4])
+    with col1:
+        if st.button('🐛', help='Show Debug Information'):
+            display_debug_info()
+    with col2:
+        st.markdown("**Show Debug Information**")
+
+    st.markdown("""---""")  # Horizontal line for visual separation
+
+    # Link to Blog Post
+    st.markdown('📖 [Blog post on setup and how the app was built](https://blog.stellersjay.pub)')
+
+    st.markdown("---")  # Horizontal line for visual separation
+    st.markdown("## 📬 Contact & Source Code")
+
+    # GitHub Repo Link
+    st.markdown("Check out the [GitHub repository](https://github.com/jwalker/your-repo-name) for this project.")
+
+    # Email Contact
+    st.markdown("Feel free to reach out via email: [jwalker](mailto:[email protected])")
+

requirements.txt

@@ -0,0 +1,58 @@
+altair==5.2.0
+attrs==23.2.0
+blinker==1.7.0
+cachetools==5.3.3
+certifi==2024.2.2
+charset-normalizer==3.3.2
+click==8.1.7
+diskcache==5.6.3
+einops==0.7.0
+filelock==3.13.1
+fsspec==2024.3.0
+gguf==0.6.0
+gitdb==4.0.11
+gitpython==3.1.42
+huggingface-hub==0.21.4
+idna==3.6
+jinja2==3.1.3
+jsonschema==4.21.1
+jsonschema-specifications==2023.12.1
+llama-cpp-python==0.2.56
+markdown-it-py==3.0.0
+markupsafe==2.1.5
+mdurl==0.1.2
+mpmath==1.3.0
+networkx==3.2.1
+numpy==1.26.4
+packaging==23.2
+pandas==2.2.1
+pillow==10.2.0
+protobuf==4.25.3
+pyarrow==15.0.2
+pydeck==0.8.0
+pygments==2.17.2
+python-dateutil==2.9.0.post0
+pytz==2024.1
+pyyaml==6.0.1
+referencing==0.34.0
+regex==2023.12.25
+requests==2.31.0
+rich==13.7.1
+rpds-py==0.18.0
+safetensors==0.4.2
+sentencepiece==0.1.99
+six==1.16.0
+smmap==5.0.1
+streamlit==1.32.2
+sympy==1.12
+tenacity==8.2.3
+tokenizers==0.15.2
+toml==0.10.2
+toolz==0.12.1
+torch==2.1.2
+tornado==6.4
+tqdm==4.66.2
+transformers==4.38.2
+typing-extensions==4.10.0
+tzdata==2024.1
+urllib3==2.2.1