UI changes

Dockerfile

@@ -2,17 +2,14 @@ FROM python:3.12-slim
 
 WORKDIR /app
 
-# install uv (fast dependency manager)
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
 RUN pip install --no-cache-dir uv
 
-# copy everything
 COPY . .
 
-# install project from pyproject.toml
 RUN uv pip install --system -e ".[dev]"
 
-# expose API port (FastAPI default)
 EXPOSE 8000
 
-# run using your pyproject entrypoint
-CMD ["server"]
+CMD ["python", "-m", "server.app"]

requirements.txt

@@ -1,6 +1,8 @@
-fastapi==0.110.0
-uvicorn==0.29.0
-pydantic==2.6.4
+fastapi>=0.110.0
+uvicorn>=0.29.0
 openai>=1.0.0
+python-dotenv>=1.0.0
 openenv-core>=0.2.0
-pyyaml>=6.0
+pydantic>=2.0.0
+gradio>=4.0.0
+requests>=2.31.0

server/app.py

@@ -14,6 +14,8 @@ import os
 import uvicorn
 from fastapi import FastAPI, HTTPException
 from pydantic import BaseModel
+from ui import build_ui
+import gradio as gr
 
 # Import your environments
 from tasks.task_simple import create_env as create_simple
@@ -108,8 +110,12 @@ def state():
     except Exception as e:
         raise HTTPException(status_code=400, detail=str(e))
 
+# Mount Gradio UI inside FastAPI
+ui = build_ui()
+app = gr.mount_gradio_app(app, ui, path="/ui")
+
 def main():
-    port = int(os.getenv("PORT", 7860))
+    port = int(os.getenv("PORT", 8000))
     uvicorn.run(app, host="0.0.0.0", port=port)
 
 

ui.py

@@ -0,0 +1,97 @@
+import gradio as gr
+import requests
+
+BASE_URL = "http://localhost:8000"
+
+current_logs = []
+current_actions = []
+
+
+def reset_env(task):
+    global current_logs, current_actions
+
+    res = requests.post(f"{BASE_URL}/reset", json={"task": task})
+    data = res.json()
+
+    current_logs = data["observation"].get("logs", [])
+    current_actions = []
+
+    return (
+        format_logs(current_logs),
+        "Environment reset",
+        "Running..."
+    )
+
+
+def step_env(action):
+    global current_logs, current_actions
+
+    res = requests.post(f"{BASE_URL}/step", json={"action": action})
+    data = res.json()
+
+    obs = data["observation"]
+    reward = data["reward"]
+    done = data["done"]
+
+    current_logs = obs.get("logs", [])
+    current_actions.append(f"{action} → reward={reward}")
+
+    status = f"{'✅ Resolved' if done else 'Running'} | Reward: {reward}"
+
+    return (
+        format_logs(current_logs),
+        "\n".join(current_actions),
+        status
+    )
+
+
+def format_logs(logs):
+    return "\n".join([f"• {log}" for log in logs])
+
+
+def build_ui():
+    with gr.Blocks() as demo:
+        gr.Markdown("#DebugOps AI Environment")
+        gr.Markdown("Interactive incident response simulator")
+
+        task_dropdown = gr.Dropdown(
+            ["simple", "multi_service", "critical"],
+            label="Select Task",
+            value="simple"
+        )
+
+        reset_btn = gr.Button("Reset Environment")
+
+        logs_box = gr.Textbox(label="Logs", lines=10)
+        actions_box = gr.Textbox(label="Actions Taken", lines=10)
+        status_box = gr.Textbox(label="Status")
+
+        with gr.Row():
+            gr.Button("Restart API").click(
+                lambda: step_env("restart_api"),
+                outputs=[logs_box, actions_box, status_box]
+            )
+            gr.Button("Restart DB").click(
+                lambda: step_env("restart_db"),
+                outputs=[logs_box, actions_box, status_box]
+            )
+            gr.Button("Restart Cache").click(
+                lambda: step_env("restart_cache"),
+                outputs=[logs_box, actions_box, status_box]
+            )
+            gr.Button("Scale Up").click(
+                lambda: step_env("scale_up"),
+                outputs=[logs_box, actions_box, status_box]
+            )
+            gr.Button("No-op").click(
+                lambda: step_env("noop"),
+                outputs=[logs_box, actions_box, status_box]
+            )
+
+        reset_btn.click(
+            reset_env,
+            inputs=task_dropdown,
+            outputs=[logs_box, actions_box, status_box]
+        )
+
+    return demo