app.py

import gradio as gr
from huggingface_hub import InferenceClient


DFIR_SYSTEM_MESSAGE = (
    "You are a DFIR and OSINT-focused assistant specializing in "
    "digital forensics, incident response, malware analysis, "
    "and investigative reasoning. Be precise, analytical, "
    "methodical, and professional. Avoid speculation."
)


def respond(
    message,
    history: list[dict[str, str]],
    dfir_mode: bool,
    system_message,
    max_tokens,
    temperature,
    top_p,
    hf_token: gr.OAuthToken,
):
    """
    Native Hugging Face inference using Gradio OAuth.
    """

    if dfir_mode:
        system_message = DFIR_SYSTEM_MESSAGE
        temperature = 0.3
        top_p = 0.9

    client = InferenceClient(
        model="mistralai/Mistral-7B-Instruct-v0.2",
        token=hf_token.token,
    )

    messages = [{"role": "system", "content": system_message}]
    messages.extend(history)
    messages.append({"role": "user", "content": message})

    response = ""

    for chunk in client.chat_completion(
        messages=messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        if chunk.choices and chunk.choices[0].delta.content:
            response += chunk.choices[0].delta.content
            yield response


chatbot = gr.ChatInterface(
    respond,
    type="messages",
    additional_inputs=[
        gr.Checkbox(
            value=True,
            label="DFIR Mode (forensic / analytical)",
        ),
        gr.Textbox(
            value="You are a helpful assistant.",
            label="System message (ignored when DFIR Mode is ON)",
        ),
        gr.Slider(1, 2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(0.1, 1.5, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-p"),
    ],
)

with gr.Blocks() as demo:
    with gr.Sidebar():
        gr.LoginButton()
    chatbot.render()

if __name__ == "__main__":
    demo.launch()