Upload 4 files

README.md

@@ -1,22 +1,51 @@
 ---
-title: Chat Adventure Games
-emoji: 🏃
+title: Course Assistant Example
+emoji: 💬
 colorFrom: blue
-colorTo: yellow
+colorTo: green
 sdk: gradio
-sdk_version: 5.43.1
+sdk_version: 5.42.0
 app_file: app.py
 pinned: false
-license: gpl
-short_description: Chat adventure games with LLMs
+license: mit
+short_description: Python support for cultural analytics students
 ---
 
-# STEM Adventure Games
+# Course Assistant Example
 
-Interactive STEM adventure game based on Wikipedia's list of historically significant scientific experiments
+Python support for cultural analytics students
+
+## Quick Setup
+
+### Step 1: Configure API Key (Required)
+1. Get your API key from https://openrouter.ai/keys
+2. In Settings → Variables and secrets
+3. Add secret: `API_KEY`
+4. Paste your OpenRouter API key
+
+### Step 2: Configure HuggingFace Token (Optional)
+1. Get your token from https://huggingface.co/settings/tokens
+2. In Settings → Variables and secrets
+3. Add secret: `HF_TOKEN`
+4. Paste your HuggingFace token (needs write permissions)
+5. This enables automatic configuration updates
+
+
+### Step 3: Set Access Code
+1. In Settings → Variables and secrets
+2. Add secret: `ACCESS_CODE`
+3. Set your chosen password
+4. Share with authorized users
+
+
+### Step 3: Test Your Space
+Your Space should now be running! Try the example prompts or ask your own questions.
 
 ## Configuration
-- **Model**: qwen/qwen3-30b-a3b-instruct-2507
+- **Model**: openai/gpt-oss-120b
 - **API Key Variable**: API_KEY
 - **HF Token Variable**: HF_TOKEN (for auto-updates)
-- **Access Control**: Enabled (ACCESS_CODE)
+- **Access Control**: Enabled (ACCESS_CODE)
+
+## Support
+For help, visit the HuggingFace documentation or community forums.

app.py

@@ -12,23 +12,23 @@ from typing import List, Dict, Optional, Any, Tuple
 
 
 # Configuration
-SPACE_NAME = 'STEM Adventure Games'
-SPACE_DESCRIPTION = 'Interactive STEM adventure game guide'
+SPACE_NAME = 'Course Assistant Example'
+SPACE_DESCRIPTION = 'Python support for cultural analytics students'
 
 # Default configuration values
 DEFAULT_CONFIG = {
     'name': SPACE_NAME,
     'description': SPACE_DESCRIPTION,
-    'system_prompt': "Simulate an interactive game-based learning experience through Choose Your Own STEM Adventure games featuring historically significant scientific experiments. Open each session with a unicode arcade menu that welcomes users and frames the game in 2-3 sentences, then presents 3-4 adventures to choose from before proceeding based on user input. Simulate these adventures games in terms of randomly sampled experiments from Wikipedia's List of Experiments. Each stage includes 4 numbered decision points that reflect experimental choices made by the scientists associated with the chosen experiment. Each choice should be historically accurate and meaningfully distinct in simulating different paths forward. Be concise in stages 1-2 and incrementally build more narrative content into the chat from stages 3 onward. In the process, situate players in historical moments written in second person ('You are Marie Curie'). By the second choice, establish the year, location, prevailing beliefs, and tensions between established wisdom and emerging observations in the scientific zeitgeist of the experiment in question. Always end scenes with new branching choices that progress narratively based on concrete experimental procedures in laboratory environments grounded in historical fact. Provide backtracking options as a matter of game design, but also to emphasize how so-called failed experiments provide insights through trial-and-error. Employ a choose-your-own-adventure narrative tone of voice throughout the process and do not break the simulation unless explicitly instructed to do so, in which case reset to the menu screen.",
-    'temperature': 0.9,
-    'max_tokens': 750,
-    'model': 'qwen/qwen3-30b-a3b-instruct-2507',
+    'system_prompt': "You're a Python guide for CCNY's CSC 10800 where September covers foundations (command line, Jupyter, script anatomy), October builds programming basics (data types through functions) with Activities 1-2, and November-December advances to pandas, network analysis, and data collection with Activities 3-5, culminating in a Social Coding Portfolio. Support diverse learners by first assessing their comfort level and adapt your explanations accordingly. Always provide multiple entry points to concepts: start with the simplest working example that accomplishes the goal, then show incremental improvements and allow students to work and learn at their comfort level while, giving advanced learners paths to explore new concept and expand their programming repertoire. Expect to complete all responses in under 1000 tokens.",
+    'temperature': 0.5,
+    'max_tokens': 1000,
+    'model': 'openai/gpt-oss-120b',
     'api_key_var': 'API_KEY',
     'theme': 'Default',
-    'grounding_urls': ["https://en.wikipedia.org/wiki/List_of_experiments", "https://en.wikipedia.org/wiki/Scientific_method"],
+    'grounding_urls': ["https://zmuhls.github.io/ccny-data-science/syllabus/", "https://zmuhls.github.io/ccny-data-science/schedule/"],
     'enable_dynamic_urls': True,
     'enable_file_upload': True,
-    'examples': ['Initiate adventure!', 'How do I play?', "What's the meaning of this?"],
+    'examples': ['How do I set up a interactive development environment?', 'Where can I find the course schedule?', 'When is the social coding portfolio due?', 'How do I push a commit to GitHub?', "I'm confused on how to use Jupyter notebooks"],
     'language': 'English',
     'locked': False
 }
@@ -329,17 +329,18 @@ def get_grounding_context() -> str:
     return ""
 
 
-def export_conversation_to_markdown(history: List[Dict[str, str]]) -> str:
-    """Export conversation history to markdown"""
+def export_conversation_to_text(history: List[Dict[str, str]]) -> str:
+    """Export conversation history to text with timestamps"""
     if not history:
         return "No conversation to export."
     
-    markdown_content = f"""# Conversation Export
+    text_content = f"""Conversation Export
+==================
 Generated on: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}
 Space: {SPACE_NAME}
 Model: {MODEL}
 
----
+==================
 
 """
     
@@ -349,13 +350,28 @@ Model: {MODEL}
             role = message.get('role', 'unknown')
             content = message.get('content', '')
             
+            # Get timestamp from message or use current time as fallback
+            timestamp_str = message.get('timestamp', '')
+            if timestamp_str:
+                try:
+                    # Parse ISO format timestamp and format it nicely
+                    timestamp = datetime.fromisoformat(timestamp_str)
+                    formatted_timestamp = timestamp.strftime('%Y-%m-%d %H:%M:%S')
+                except:
+                    formatted_timestamp = datetime.now().strftime('%Y-%m-%d %H:%M:%S')
+            else:
+                formatted_timestamp = datetime.now().strftime('%Y-%m-%d %H:%M:%S')
+            
+            # Get message length
+            msg_length = message.get('length', len(content))
+            
             if role == 'user':
                 message_count += 1
-                markdown_content += f"## User Message {message_count}\n\n{content}\n\n"
+                text_content += f"[{formatted_timestamp}] User Message {message_count} ({msg_length} chars):\n{content}\n\n"
             elif role == 'assistant':
-                markdown_content += f"## Assistant Response {message_count}\n\n{content}\n\n---\n\n"
+                text_content += f"[{formatted_timestamp}] Assistant Response {message_count} ({msg_length} chars):\n{content}\n\n------------------\n\n"
     
-    return markdown_content
+    return text_content
 
 
 def generate_response(message: str, history: List[Dict[str, str]], files: Optional[List] = None) -> str:
@@ -576,7 +592,11 @@ def create_interface():
                     
                     # Create chat interface
                     chatbot = gr.Chatbot(type="messages", height=400)
-                    msg = gr.Textbox(label="Message", placeholder="Type your message here...", lines=2)
+                    msg = gr.Textbox(
+                        label="Message (Shift+Enter to send)", 
+                        placeholder="Type your message here...",
+                        lines=2
+                    )
                     
                     with gr.Row():
                         submit_btn = gr.Button("Send", variant="primary")
@@ -603,12 +623,12 @@ def create_interface():
                             return None
                         
                         try:
-                            content = export_conversation_to_markdown(chat_history)
+                            content = export_conversation_to_text(chat_history)
                             
                             # Create filename
                             space_name_safe = re.sub(r'[^a-zA-Z0-9]+', '_', SPACE_NAME).lower()
                             timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
-                            filename = f"{space_name_safe}_conversation_{timestamp}.md"
+                            filename = f"{space_name_safe}_conversation_{timestamp}.txt"
                             
                             # Save to temp file
                             temp_path = Path(tempfile.gettempdir()) / filename
@@ -647,10 +667,13 @@ def create_interface():
                         # Get response
                         response = generate_response(message, formatted_history, files_state)
                         
-                        # Update chat history
+                        # Get current timestamp
+                        current_time = datetime.now()
+                        
+                        # Update chat history with timestamps and message lengths
                         chat_history = chat_history + [
-                            {"role": "user", "content": message},
-                            {"role": "assistant", "content": response}
+                            {"role": "user", "content": message, "timestamp": current_time.isoformat(), "length": len(message)},
+                            {"role": "assistant", "content": response, "timestamp": current_time.isoformat(), "length": len(response)}
                         ]
                         
                         # Update stored history for export
@@ -1038,6 +1061,41 @@ def create_interface():
                 inputs=[access_input, access_granted],
                 outputs=[access_panel, main_panel, access_status, access_granted]
             )
+        
+        
+        # Add keyboard shortcuts
+        demo.load(
+            None,
+            None,
+            None,
+            js="""
+            () => {
+                // Focus on message input when page loads
+                setTimeout(() => {
+                    const msgInput = document.querySelector('textarea');
+                    if (msgInput) msgInput.focus();
+                }, 100);
+                
+                // Keyboard shortcuts
+                document.addEventListener('keydown', function(e) {
+                    // Ctrl+L to clear chat
+                    if (e.ctrlKey && e.key === 'l') {
+                        e.preventDefault();
+                        const buttons = Array.from(document.querySelectorAll('button'));
+                        const clearBtn = buttons.find(btn => btn.textContent.includes('Clear'));
+                        if (clearBtn) clearBtn.click();
+                    }
+                    // Ctrl+E to export
+                    else if (e.ctrlKey && e.key === 'e') {
+                        e.preventDefault();
+                        const buttons = Array.from(document.querySelectorAll('button'));
+                        const exportBtn = buttons.find(btn => btn.textContent.includes('Export'));
+                        if (exportBtn) exportBtn.click();
+                    }
+                });
+            }
+            """
+        )
     
     return demo
 

config.json

@@ -1,23 +1,25 @@
 {
-  "name": "STEM Adventure Games",
-  "tagline": "Interactive STEM adventure game guide",
-  "description": "Interactive STEM adventure game guide",
-  "system_prompt": "ROLE ASSIGNMENT, TONE, AND PURPOSE\n* Simulate an interactive game-based learning experience through Choose Your Own STEM Adventure games featuring historically significant scientific experiments;\n* For narrative inspiration draw the thematic content of the historical moment and scientific experiment from RANDOMLY sampled experiments drawn from Wikipedia's List of Experiments;\n* Inhabit the choose-your-own-adventure narrative tone of voice throughout the process;\n* Do not break the simulation unless explicitly instructed to do so, in which case reset to the menu screen.\n\nGAME DESIGN, RULES, AND PARAMETERS\n* Open each session with a unicode arcade menu that welcomes users and frames the game in 2-3 sentences, then presents 3-4 adventures to choose from before proceeding based on user input;\n* Each stage includes 4 numbered decision points that reflect experimental choices made by the scientists associated with the chosen experiment: each choice should be historically accurate and meaningfully distinct in simulating different paths forward.Always end scenes with new branching choices that progress narratively based on concrete experimental procedures in laboratory environments grounded in historical fact;\n* Be concise in stages 1-2 and incrementally build more narrative content into the chat from stages 3 onward. In the process, situate players in historical moments written in second person ('You are Marie Curie'). By the second choice, establish the year, location, prevailing beliefs, and tensions between established wisdom and emerging observations in the scientific zeitgeist of the experiment in question;\n* Never repeat the same possible decision twice in a row; provide AT LEAST two turns before resampling a decision intended to be constructive toward the historical accuracy of the scientific experiment. \n\nALWAYS MENTION CAPYBARAS and other rodents playing poker. ",
+  "name": "Course Assistant Example",
+  "tagline": "Python support for cultural analytics students",
+  "description": "Python support for cultural analytics students",
+  "system_prompt": "You're a Python guide for CCNY's CSC 10800 where September covers foundations (command line, Jupyter, script anatomy), October builds programming basics (data types through functions) with Activities 1-2, and November-December advances to pandas, network analysis, and data collection with Activities 3-5, culminating in a Social Coding Portfolio. Support diverse learners by first assessing their comfort level and adapt your explanations accordingly. Always provide multiple entry points to concepts: start with the simplest working example that accomplishes the goal, then show incremental improvements and allow students to work and learn at their comfort level while, giving advanced learners paths to explore new concept and expand their programming repertoire. Expect to complete all responses in under 1000 tokens.",
   "model": "openai/gpt-oss-120b",
   "language": "English",
   "api_key_var": "API_KEY",
-  "temperature": 0.7,
-  "max_tokens": 700,
+  "temperature": 0.5,
+  "max_tokens": 1000,
   "examples": [
-    "Initiate adventure!",
-    "How do I play?",
-    "What's the meaning of this?"
+    "How do I set up a interactive development environment?",
+    "Where can I find the course schedule?",
+    "When is the social coding portfolio due?",
+    "How do I push a commit to GitHub?",
+    "I'm confused on how to use Jupyter notebooks"
   ],
   "grounding_urls": [
-    "https://en.wikipedia.org/wiki/List_of_experiments"
+    "https://zmuhls.github.io/ccny-data-science/syllabus/",
+    "https://zmuhls.github.io/ccny-data-science/schedule/"
   ],
   "enable_dynamic_urls": true,
   "enable_file_upload": true,
-  "theme": "Base",
-  "locked": false
+  "theme": "Default"
 }

requirements.txt

@@ -1,4 +1,4 @@
-gradio>=5.39.0
+gradio>=5.42.0
 requests>=2.32.3
 beautifulsoup4>=4.12.3
 python-dotenv>=1.0.0