- **Assistant**: Display name of the assistant

- **Website**: Link to the assistant's homepage or documentation

README.md

@@ -28,6 +28,8 @@ If an assistant can consistently get pull requests accepted across different pro
 Key metrics from the last 180 days:
 
 **Leaderboard Table**
+- **Assistant**: Display name of the assistant
+- **Website**: Link to the assistant's homepage or documentation
 - **Total PRs**: Pull requests the assistant has opened
 - **Merged PRs**: PRs that got merged (not just closed)
 - **Acceptance Rate**: Percentage of concluded PRs that got merged

app.py

@@ -22,13 +22,13 @@ load_dotenv()
 # CONFIGURATION
 # =============================================================================
 
-AGENTS_REPO = "SWE-Arena/bot_metadata"  # HuggingFace dataset for agent metadata
+AGENTS_REPO = "SWE-Arena/bot_metadata"  # HuggingFace dataset for assistant metadata
 LEADERBOARD_FILENAME = f"{os.getenv('COMPOSE_PROJECT_NAME')}.json"
 LEADERBOARD_REPO = "SWE-Arena/leaderboard_data"  # HuggingFace dataset for leaderboard data
 MAX_RETRIES = 5
 
 LEADERBOARD_COLUMNS = [
-    ("Agent Name", "string"),
+    ("Assistant", "string"),
     ("Website", "string"),
     ("Total PRs", "number"),
     ("Merged PRs", "number"),
@@ -96,10 +96,10 @@ def validate_github_username(identifier):
 # =============================================================================
 
 def load_agents_from_hf():
-    """Load all agent metadata JSON files from HuggingFace dataset."""
+    """Load all assistant metadata JSON files from HuggingFace dataset."""
     try:
         api = HfApi()
-        agents = []
+        assistants = []
 
         # List all files in the repository
         files = list_repo_files_with_backoff(api=api, repo_id=AGENTS_REPO, repo_type="dataset")
@@ -119,27 +119,27 @@ def load_agents_from_hf():
                 with open(file_path, 'r') as f:
                     agent_data = json.load(f)
 
-                    # Only process agents with status == "active"
+                    # Only process assistants with status == "active"
                     if agent_data.get('status') != 'active':
                         continue
 
-                    # Extract github_identifier from filename (e.g., "agent[bot].json" -> "agent[bot]")
+                    # Extract github_identifier from filename (e.g., "assistant[bot].json" -> "assistant[bot]")
                     filename_identifier = json_file.replace('.json', '')
 
                     # Add or override github_identifier to match filename
                     agent_data['github_identifier'] = filename_identifier
 
-                    agents.append(agent_data)
+                    assistants.append(agent_data)
 
             except Exception as e:
                 print(f"Warning: Could not load {json_file}: {str(e)}")
                 continue
 
-        print(f"Loaded {len(agents)} agents from HuggingFace")
-        return agents
+        print(f"Loaded {len(assistants)} assistants from HuggingFace")
+        return assistants
 
     except Exception as e:
-        print(f"Could not load agents from HuggingFace: {str(e)}")
+        print(f"Could not load assistants from HuggingFace: {str(e)}")
         return None
 
 
@@ -195,7 +195,7 @@ def upload_with_retry(api, path_or_fileobj, path_in_repo, repo_id, repo_type, to
 
 
 def save_agent_to_hf(data):
-    """Save a new agent to HuggingFace dataset as {identifier}.json in root."""
+    """Save a new assistant to HuggingFace dataset as {identifier}.json in root."""
     try:
         api = HfApi()
         token = get_hf_token()
@@ -220,7 +220,7 @@ def save_agent_to_hf(data):
                 repo_type="dataset",
                 token=token
             )
-            print(f"Saved agent to HuggingFace: {filename}")
+            print(f"Saved assistant to HuggingFace: {filename}")
             return True
         finally:
             # Always clean up local file, even if upload fails
@@ -228,7 +228,7 @@ def save_agent_to_hf(data):
                 os.remove(filename)
 
     except Exception as e:
-        print(f"Error saving agent: {str(e)}")
+        print(f"Error saving assistant: {str(e)}")
         return False
 
 
@@ -275,10 +275,10 @@ def create_monthly_metrics_plot(top_n=5):
     - Left y-axis: Acceptance Rate (%) as line curves
     - Right y-axis: Total PRs created as bar charts
 
-    Each agent gets a unique color for both their line and bars.
+    Each assistant gets a unique color for both their line and bars.
 
     Args:
-        top_n: Number of top agents to show (default: 5)
+        top_n: Number of top assistants to show (default: 5)
     """
     # Load from saved dataset
     saved_data = load_leaderboard_data_from_hf()
@@ -303,10 +303,10 @@ def create_monthly_metrics_plot(top_n=5):
     print(f"Loaded monthly metrics from saved dataset")
 
     # Apply top_n filter if specified
-    if top_n is not None and top_n > 0 and metrics.get('agents'):
-        # Calculate total PRs for each agent
+    if top_n is not None and top_n > 0 and metrics.get('assistants'):
+        # Calculate total PRs for each assistant
         agent_totals = []
-        for agent_name in metrics['agents']:
+        for agent_name in metrics['assistants']:
             agent_data = metrics['data'].get(agent_name, {})
             total_prs = sum(agent_data.get('total_prs', []))
             agent_totals.append((agent_name, total_prs))
@@ -315,14 +315,14 @@ def create_monthly_metrics_plot(top_n=5):
         agent_totals.sort(key=lambda x: x[1], reverse=True)
         top_agents = [agent_name for agent_name, _ in agent_totals[:top_n]]
 
-        # Filter metrics to only include top agents
+        # Filter metrics to only include top assistants
         metrics = {
-            'agents': top_agents,
+            'assistants': top_agents,
             'months': metrics['months'],
-            'data': {agent: metrics['data'][agent] for agent in top_agents if agent in metrics['data']}
+            'data': {assistant: metrics['data'][assistant] for assistant in top_agents if assistant in metrics['data']}
         }
 
-    if not metrics['agents'] or not metrics['months']:
+    if not metrics['assistants'] or not metrics['months']:
         # Return an empty figure with a message
         fig = go.Figure()
         fig.add_annotation(
@@ -341,7 +341,7 @@ def create_monthly_metrics_plot(top_n=5):
     # Create figure with secondary y-axis
     fig = make_subplots(specs=[[{"secondary_y": True}]])
 
-    # Generate unique colors for many agents using HSL color space
+    # Generate unique colors for many assistants using HSL color space
     def generate_color(index, total):
         """Generate distinct colors using HSL color space for better distribution"""
         hue = (index * 360 / total) % 360
@@ -349,15 +349,15 @@ def create_monthly_metrics_plot(top_n=5):
         lightness = 45 + (index % 2) * 10   # Vary lightness slightly
         return f'hsl({hue}, {saturation}%, {lightness}%)'
 
-    agents = metrics['agents']
+    assistants = metrics['assistants']
     months = metrics['months']
     data = metrics['data']
 
-    # Generate colors for all agents
-    agent_colors = {agent: generate_color(idx, len(agents)) for idx, agent in enumerate(agents)}
+    # Generate colors for all assistants
+    agent_colors = {assistant: generate_color(idx, len(assistants)) for idx, assistant in enumerate(assistants)}
 
-    # Add traces for each agent
-    for idx, agent_name in enumerate(agents):
+    # Add traces for each assistant
+    for idx, agent_name in enumerate(assistants):
         color = agent_colors[agent_name]
         agent_data = data[agent_name]
 
@@ -377,8 +377,8 @@ def create_monthly_metrics_plot(top_n=5):
                     line=dict(color=color, width=2),
                     marker=dict(size=8),
                     legendgroup=agent_name,
-                    showlegend=(top_n is not None and top_n <= 10),  # Show legend for top N agents
-                    hovertemplate='<b>Agent: %{fullData.name}</b><br>' +
+                    showlegend=(top_n is not None and top_n <= 10),  # Show legend for top N assistants
+                    hovertemplate='<b>Assistant: %{fullData.name}</b><br>' +
                                  'Month: %{x}<br>' +
                                  'Acceptance Rate: %{y:.2f}%<br>' +
                                  '<extra></extra>'
@@ -387,7 +387,7 @@ def create_monthly_metrics_plot(top_n=5):
             )
 
         # Add bar trace for total PRs (right y-axis)
-        # Only show bars for months where agent has PRs
+        # Only show bars for months where assistant has PRs
         x_bars = []
         y_bars = []
         for month, count in zip(months, agent_data['total_prs']):
@@ -404,11 +404,11 @@ def create_monthly_metrics_plot(top_n=5):
                     marker=dict(color=color, opacity=0.6),
                     legendgroup=agent_name,
                     showlegend=False,  # Hide duplicate legend entry (already shown in Scatter)
-                    hovertemplate='<b>Agent: %{fullData.name}</b><br>' +
+                    hovertemplate='<b>Assistant: %{fullData.name}</b><br>' +
                                  'Month: %{x}<br>' +
                                  'Total PRs: %{y}<br>' +
                                  '<extra></extra>',
-                    offsetgroup=agent_name  # Group bars by agent for proper spacing
+                    offsetgroup=agent_name  # Group bars by assistant for proper spacing
                 ),
                 secondary_y=True
             )
@@ -430,7 +430,7 @@ def create_monthly_metrics_plot(top_n=5):
     show_legend = (top_n is not None and top_n <= 10)
     fig.update_layout(
         title=None,
-        hovermode='closest',  # Show individual agent info on hover
+        hovermode='closest',  # Show individual assistant info on hover
         barmode='group',
         height=600,
         showlegend=show_legend,
@@ -469,9 +469,9 @@ def get_leaderboard_dataframe():
     filtered_count = 0
     for identifier, data in cache_dict.items():
         total_prs = data.get('total_prs', 0)
-        print(f"   Agent '{identifier}': {total_prs} PRs")
+        print(f"   Assistant '{identifier}': {total_prs} PRs")
 
-        # Filter out agents with zero total PRs
+        # Filter out assistants with zero total PRs
         if total_prs == 0:
             filtered_count += 1
             continue
@@ -485,8 +485,8 @@ def get_leaderboard_dataframe():
             data.get('acceptance_rate', 0.0),
         ])
 
-    print(f"Filtered out {filtered_count} agents with 0 PRs")
-    print(f"Leaderboard will show {len(rows)} agents")
+    print(f"Filtered out {filtered_count} assistants with 0 PRs")
+    print(f"Leaderboard will show {len(rows)} assistants")
 
     # Create DataFrame
     column_names = [col[0] for col in LEADERBOARD_COLUMNS]
@@ -510,14 +510,14 @@ def get_leaderboard_dataframe():
 
 def submit_agent(identifier, agent_name, organization, website):
     """
-    Submit a new agent to the leaderboard.
+    Submit a new assistant to the leaderboard.
     Validates input and saves submission.
     """
     # Validate required fields
     if not identifier or not identifier.strip():
         return "ERROR: GitHub identifier is required", gr.update()
     if not agent_name or not agent_name.strip():
-        return "ERROR: Agent name is required", gr.update()
+        return "ERROR: Assistant name is required", gr.update()
     if not organization or not organization.strip():
         return "ERROR: Organization name is required", gr.update()
     if not website or not website.strip():
@@ -534,12 +534,12 @@ def submit_agent(identifier, agent_name, organization, website):
     if not is_valid:
         return f"ERROR: {message}", gr.update()
 
-    # Check for duplicates by loading agents from HuggingFace
-    agents = load_agents_from_hf()
-    if agents:
-        existing_names = {agent['github_identifier'] for agent in agents}
+    # Check for duplicates by loading assistants from HuggingFace
+    assistants = load_agents_from_hf()
+    if assistants:
+        existing_names = {assistant['github_identifier'] for assistant in assistants}
         if identifier in existing_names:
-            return f"WARNING: Agent with identifier '{identifier}' already exists", gr.update()
+            return f"WARNING: Assistant with identifier '{identifier}' already exists", gr.update()
 
     # Create submission
     submission = {
@@ -576,7 +576,7 @@ def reload_leaderboard_data():
         if data:
             print(f"Successfully reloaded leaderboard data")
             print(f"   Last updated: {data.get('metadata', {}).get('last_updated', 'Unknown')}")
-            print(f"   Agents: {len(data.get('leaderboard', {}))}")
+            print(f"   Assistants: {len(data.get('leaderboard', {}))}")
         else:
             print(f"No data available")
     except Exception as e:
@@ -589,7 +589,7 @@ def reload_leaderboard_data():
 # GRADIO APPLICATION
 # =============================================================================
 
-print(f"\nStarting SWE Agent PR Leaderboard")
+print(f"\nStarting SWE Assistant PR Leaderboard")
 print(f"   Data source: {LEADERBOARD_REPO}")
 print(f"   Reload frequency: Daily at 12:00 AM UTC\n")
 
@@ -610,19 +610,19 @@ print(f"On startup: Loads cached data from HuggingFace on demand")
 print(f"{'='*80}\n")
 
 # Create Gradio interface
-with gr.Blocks(title="SWE Agent PR Leaderboard", theme=gr.themes.Soft()) as app:
-    gr.Markdown("# SWE Agent PR Leaderboard")
-    gr.Markdown(f"Track and compare GitHub pull request statistics for SWE agents")
+with gr.Blocks(title="SWE Assistant PR Leaderboard", theme=gr.themes.Soft()) as app:
+    gr.Markdown("# SWE Assistant PR Leaderboard")
+    gr.Markdown(f"Track and compare GitHub pull request statistics for SWE assistants")
 
     with gr.Tabs():
 
         # Leaderboard Tab
         with gr.Tab("Leaderboard"):
-            gr.Markdown("*Statistics are based on agent PR activity tracked by the system*")
+            gr.Markdown("*Statistics are based on assistant PR activity tracked by the system*")
             leaderboard_table = Leaderboard(
                 value=pd.DataFrame(columns=[col[0] for col in LEADERBOARD_COLUMNS]),  # Empty initially
                 datatype=LEADERBOARD_COLUMNS,
-                search_columns=["Agent Name", "Website"],
+                search_columns=["Assistant", "Website"],
                 filter_columns=[
                     ColumnFilter(
                         "Acceptance Rate (%)",
@@ -645,8 +645,8 @@ with gr.Blocks(title="SWE Agent PR Leaderboard", theme=gr.themes.Soft()) as app:
             # Monthly Metrics Section
             gr.Markdown("---")  # Divider
             with gr.Group():
-                gr.Markdown("### Monthly Performance - Top 5 Agents")
-                gr.Markdown("*Shows acceptance rate trends and PR volumes for the most active agents*")
+                gr.Markdown("### Monthly Performance - Top 5 Assistants")
+                gr.Markdown("*Shows acceptance rate trends and PR volumes for the most active assistants*")
                 monthly_metrics_plot = gr.Plot(label="Monthly Metrics")
 
             # Load monthly metrics when app starts
@@ -657,20 +657,20 @@ with gr.Blocks(title="SWE Agent PR Leaderboard", theme=gr.themes.Soft()) as app:
             )
 
 
-        # Submit Agent Tab
-        with gr.Tab("Submit Your Agent"):
+        # Submit Assistant Tab
+        with gr.Tab("Submit Your Assistant"):
 
-            gr.Markdown("Fill in the details below to add your agent to the leaderboard.")
+            gr.Markdown("Fill in the details below to add your assistant to the leaderboard.")
 
             with gr.Row():
                 with gr.Column():
                     github_input = gr.Textbox(
                         label="GitHub Identifier*",
-                        placeholder="Your agent username (e.g., my-agent[bot])"
+                        placeholder="Your assistant username (e.g., my-assistant[bot])"
                     )
                     name_input = gr.Textbox(
-                        label="Agent Name*",
-                        placeholder="Your agent's display name"
+                        label="Assistant Name*",
+                        placeholder="Your assistant's display name"
                     )
 
                 with gr.Column():
@@ -680,11 +680,11 @@ with gr.Blocks(title="SWE Agent PR Leaderboard", theme=gr.themes.Soft()) as app:
                     )
                     website_input = gr.Textbox(
                         label="Website*",
-                        placeholder="https://your-agent-website.com"
+                        placeholder="https://your-assistant-website.com"
                     )
 
             submit_button = gr.Button(
-                "Submit Agent",
+                "Submit Assistant",
                 variant="primary"
             )
             submission_status = gr.Textbox(

msr.py

@@ -363,7 +363,7 @@ def fetch_all_pr_metadata_streaming(conn, identifiers, start_date, end_date):
         end_date: End datetime (timezone-aware)
     
     Returns:
-        Dictionary mapping agent identifier to list of PR metadata
+        Dictionary mapping assistant identifier to list of PR metadata
     """
     identifier_list = ', '.join([f"'{id}'" for id in identifiers])
     metadata_by_agent = defaultdict(list)
@@ -495,7 +495,7 @@ def fetch_all_pr_metadata_streaming(conn, identifiers, start_date, end_date):
     
     # Final summary
     agents_with_data = sum(1 for prs in metadata_by_agent.values() if prs)
-    print(f"\n   ✓ Complete: {total_prs} PRs found for {agents_with_data}/{len(identifiers)} agents")
+    print(f"\n   ✓ Complete: {total_prs} PRs found for {agents_with_data}/{len(identifiers)} assistants")
     
     return dict(metadata_by_agent)
 
@@ -558,14 +558,14 @@ def sync_agents_repo():
 
 def load_agents_from_hf():
     """
-    Load all agent metadata JSON files from local git repository.
+    Load all assistant metadata JSON files from local git repository.
     ALWAYS syncs with remote first to ensure we have the latest bot data.
     """
     # MANDATORY: Sync with remote first to get latest bot data
-    print(f"   Syncing bot_data repository to get latest agents...")
+    print(f"   Syncing bot_data repository to get latest assistants...")
     sync_agents_repo()  # Will raise exception if sync fails
 
-    agents = []
+    assistants = []
 
     # Scan local directory for JSON files
     if not os.path.exists(AGENTS_REPO_LOCAL_PATH):
@@ -573,7 +573,7 @@ def load_agents_from_hf():
 
     # Walk through the directory to find all JSON files
     files_processed = 0
-    print(f"   Loading agent metadata from {AGENTS_REPO_LOCAL_PATH}...")
+    print(f"   Loading assistant metadata from {AGENTS_REPO_LOCAL_PATH}...")
 
     for root, dirs, files in os.walk(AGENTS_REPO_LOCAL_PATH):
         # Skip .git directory
@@ -591,7 +591,7 @@ def load_agents_from_hf():
                 with open(file_path, 'r', encoding='utf-8') as f:
                     agent_data = json.load(f)
 
-                # Only include active agents
+                # Only include active assistants
                 if agent_data.get('status') != 'active':
                     continue
 
@@ -599,14 +599,14 @@ def load_agents_from_hf():
                 github_identifier = filename.replace('.json', '')
                 agent_data['github_identifier'] = github_identifier
 
-                agents.append(agent_data)
+                assistants.append(agent_data)
 
             except Exception as e:
                 print(f"   ⚠ Error loading {filename}: {str(e)}")
                 continue
 
-    print(f"   ✓ Loaded {len(agents)} active agents (from {files_processed} total files)")
-    return agents
+    print(f"   ✓ Loaded {len(assistants)} active assistants (from {files_processed} total files)")
+    return assistants
 
 
 def calculate_pr_stats_from_metadata(metadata_list):
@@ -626,12 +626,12 @@ def calculate_pr_stats_from_metadata(metadata_list):
     }
 
 
-def calculate_monthly_metrics_by_agent(all_metadata_dict, agents):
-    """Calculate monthly metrics for all agents for visualization."""
-    identifier_to_name = {agent.get('github_identifier'): agent.get('name') for agent in agents if agent.get('github_identifier')}
+def calculate_monthly_metrics_by_agent(all_metadata_dict, assistants):
+    """Calculate monthly metrics for all assistants for visualization."""
+    identifier_to_name = {assistant.get('github_identifier'): assistant.get('name') for assistant in assistants if assistant.get('github_identifier')}
 
     if not all_metadata_dict:
-        return {'agents': [], 'months': [], 'data': {}}
+        return {'assistants': [], 'months': [], 'data': {}}
 
     agent_month_data = defaultdict(lambda: defaultdict(list))
 
@@ -690,30 +690,30 @@ def calculate_monthly_metrics_by_agent(all_metadata_dict, agents):
     agents_list = sorted(list(agent_month_data.keys()))
 
     return {
-        'agents': agents_list,
+        'assistants': agents_list,
         'months': months,
         'data': result_data
     }
 
 
-def construct_leaderboard_from_metadata(all_metadata_dict, agents):
+def construct_leaderboard_from_metadata(all_metadata_dict, assistants):
     """Construct leaderboard from in-memory PR metadata."""
-    if not agents:
-        print("Error: No agents found")
+    if not assistants:
+        print("Error: No assistants found")
         return {}
 
     cache_dict = {}
 
-    for agent in agents:
-        identifier = agent.get('github_identifier')
-        agent_name = agent.get('name', 'Unknown')
+    for assistant in assistants:
+        identifier = assistant.get('github_identifier')
+        agent_name = assistant.get('name', 'Unknown')
 
         bot_metadata = all_metadata_dict.get(identifier, [])
         stats = calculate_pr_stats_from_metadata(bot_metadata)
 
         cache_dict[identifier] = {
             'name': agent_name,
-            'website': agent.get('website', 'N/A'),
+            'website': assistant.get('website', 'N/A'),
             'github_identifier': identifier,
             **stats
         }
@@ -767,7 +767,7 @@ def save_leaderboard_data_to_hf(leaderboard_dict, monthly_metrics):
 
 def mine_all_agents():
     """
-    Mine PR metadata for all agents using STREAMING batch processing.
+    Mine PR metadata for all assistants using STREAMING batch processing.
     Downloads GHArchive data, then uses BATCH-based DuckDB queries.
     """
     print(f"\n[1/4] Downloading GHArchive data...")
@@ -775,19 +775,19 @@ def mine_all_agents():
     if not download_all_gharchive_data():
         print("Warning: Download had errors, continuing with available data...")
 
-    print(f"\n[2/4] Loading agent metadata...")
+    print(f"\n[2/4] Loading assistant metadata...")
 
-    agents = load_agents_from_hf()
-    if not agents:
-        print("Error: No agents found")
+    assistants = load_agents_from_hf()
+    if not assistants:
+        print("Error: No assistants found")
         return
 
-    identifiers = [agent['github_identifier'] for agent in agents if agent.get('github_identifier')]
+    identifiers = [assistant['github_identifier'] for assistant in assistants if assistant.get('github_identifier')]
     if not identifiers:
-        print("Error: No valid agent identifiers found")
+        print("Error: No valid assistant identifiers found")
         return
 
-    print(f"\n[3/4] Mining PR metadata ({len(identifiers)} agents, {LEADERBOARD_TIME_FRAME_DAYS} days)...")
+    print(f"\n[3/4] Mining PR metadata ({len(identifiers)} assistants, {LEADERBOARD_TIME_FRAME_DAYS} days)...")
 
     try:
         conn = get_duckdb_connection()
@@ -815,8 +815,8 @@ def mine_all_agents():
     print(f"\n[4/4] Saving leaderboard...")
 
     try:
-        leaderboard_dict = construct_leaderboard_from_metadata(all_metadata, agents)
-        monthly_metrics = calculate_monthly_metrics_by_agent(all_metadata, agents)
+        leaderboard_dict = construct_leaderboard_from_metadata(all_metadata, assistants)
+        monthly_metrics = calculate_monthly_metrics_by_agent(all_metadata, assistants)
         save_leaderboard_data_to_hf(leaderboard_dict, monthly_metrics)
 
     except Exception as e:
@@ -850,7 +850,7 @@ def setup_scheduler():
         mine_all_agents,
         trigger=trigger,
         id='mine_all_agents',
-        name='Mine GHArchive data for all agents',
+        name='Mine GHArchive data for all assistants',
         replace_existing=True
     )