app.py

import gradio as gr
import asyncio
import json
import logging
import traceback
from agentic_reliability_framework.runtime.engine import EnhancedReliabilityEngine

# Configure logging to show details
logging.basicConfig(level=logging.DEBUG, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
logger = logging.getLogger(__name__)

# Initialize the engine
try:
    logger.info("Initializing EnhancedReliabilityEngine...")
    engine = EnhancedReliabilityEngine()
    logger.info("Engine initialized successfully.")
except Exception as e:
    logger.error(f"Failed to initialize engine: {e}\n{traceback.format_exc()}")
    engine = None

async def analyze(component, latency, error_rate, throughput, cpu_util, memory_util):
    """Call the ARF v4 engine with telemetry data."""
    if engine is None:
        return json.dumps({"error": "Engine failed to initialize. Check logs."}, indent=2)
    try:
        logger.info(f"Analyzing: component={component}, latency={latency}, error_rate={error_rate}, throughput={throughput}, cpu={cpu_util}, mem={memory_util}")
        result = await engine.process_event_enhanced(
            component=component,
            latency=float(latency),
            error_rate=float(error_rate),
            throughput=float(throughput) if throughput else 1000.0,
            cpu_util=float(cpu_util) if cpu_util else None,
            memory_util=float(memory_util) if memory_util else None
        )
        logger.info("Analysis completed successfully.")
        return json.dumps(result, indent=2)
    except Exception as e:
        logger.error(f"Error during analysis: {e}\n{traceback.format_exc()}")
        return json.dumps({"error": str(e), "traceback": traceback.format_exc()}, indent=2)

def sync_analyze(*args):
    """Synchronous wrapper for Gradio."""
    return asyncio.run(analyze(*args))

# Define the Gradio interface
with gr.Blocks(title="ARF v4 – Reliability Lab", theme="soft") as demo:
    gr.Markdown("""
    # 🧠 Agentic Reliability Framework v4
    **Hybrid Bayesian + HMC intelligence for infrastructure reliability**
    
    Enter telemetry below to see ARF's advisory analysis. All outputs are **OSS advisory only** – no execution.
    """)
    
    with gr.Row():
        with gr.Column():
            component = gr.Dropdown(
                choices=["api-service", "auth-service", "payment-service", "database", "cache-service"],
                value="api-service",
                label="Component"
            )
            latency = gr.Slider(10, 1000, value=100, label="Latency P99 (ms)")
            error_rate = gr.Slider(0, 0.5, value=0.02, step=0.001, label="Error Rate")
            throughput = gr.Number(value=1000, label="Throughput (req/s)")
            cpu_util = gr.Slider(0, 1, value=0.4, label="CPU Utilization")
            memory_util = gr.Slider(0, 1, value=0.3, label="Memory Utilization")
            submit = gr.Button("🚀 Analyze", variant="primary")
        
        with gr.Column():
            output = gr.JSON(label="ARF Analysis Result")
    
    submit.click(
        fn=sync_analyze,
        inputs=[component, latency, error_rate, throughput, cpu_util, memory_util],
        outputs=output
    )
    
    gr.Markdown("""
    ---
    ### 📚 About This Demo
    - Uses the full **ARF v4 engine** (`EnhancedReliabilityEngine`)
    - Risk scores combine **online conjugate priors** + **offline HMC** (if trained)
    - Multi‑agent system runs in parallel (detective, diagnostician, predictive)
    - Optional Claude synthesis (if `ANTHROPIC_API_KEY` is set)
    
    [📖 Tutorial](https://github.com/petter2025us/agentic-reliability-framework/blob/main/TUTORIAL.md) | 
    [🐙 GitHub](https://github.com/petter2025us/agentic-reliability-framework) |
    [💼 Enterprise](mailto:petter2025us@outlook.com)
    """)

if __name__ == "__main__":
    demo.launch(server_name="0.0.0.0", server_port=7860)