diff --git a/benchmarks/.DS_Store b/benchmarks/.DS_Store
index 15b5bd44..b4d6cfce 100644
Binary files a/benchmarks/.DS_Store and b/benchmarks/.DS_Store differ
diff --git a/control-plane/src/components/documents-view.tsx b/control-plane/src/components/documents-view.tsx
index 630e344d..f209855d 100644
--- a/control-plane/src/components/documents-view.tsx
+++ b/control-plane/src/components/documents-view.tsx
@@ -62,7 +62,7 @@ export function DocumentsView() {
         type="text"
         value={searchQuery}
         onChange={(e) => setSearchQuery(e.target.value)}
-        placeholder="Search documents (ID, metadata)..."
+        placeholder="Search documents (ID)..."
         className="w-full max-w-2xl px-2.5 py-2 mb-4 mx-5 border-2 border-border bg-background text-foreground rounded text-sm focus:outline-none focus:ring-2 focus:ring-ring"
       />
 
@@ -75,7 +75,6 @@ export function DocumentsView() {
               <th className="p-2.5 text-left border border-border bg-card text-card-foreground">Updated</th>
               <th className="p-2.5 text-left border border-border bg-card text-card-foreground">Text Length</th>
               <th className="p-2.5 text-left border border-border bg-card text-card-foreground">Memory Units</th>
-              <th className="p-2.5 text-left border border-border bg-card text-card-foreground">Metadata</th>
               <th className="p-2.5 text-left border border-border bg-card text-card-foreground">Actions</th>
             </tr>
           </thead>
@@ -94,11 +93,6 @@ export function DocumentsView() {
                   </td>
                   <td className="p-2 border border-border">{doc.text_length?.toLocaleString()} chars</td>
                   <td className="p-2 border border-border">{doc.memory_unit_count}</td>
-                  <td className="p-2 border border-border" title={JSON.stringify(doc.metadata)}>
-                    {Object.keys(doc.metadata || {}).length > 0
-                      ? JSON.stringify(doc.metadata).substring(0, 50) + '...'
-                      : 'None'}
-                  </td>
                   <td className="p-2 border border-border">
                     <button
                       onClick={() => viewDocumentText(doc.id)}
@@ -112,7 +106,7 @@ export function DocumentsView() {
               ))
             ) : (
               <tr>
-                <td colSpan={7} className="p-10 text-center text-muted-foreground bg-muted">
+                <td colSpan={6} className="p-10 text-center text-muted-foreground bg-muted">
                   Click "Load Documents" to view data
                 </td>
               </tr>
diff --git a/memora-cli/README.md b/memora-cli/README.md
new file mode 100644
index 00000000..250ccb69
--- /dev/null
+++ b/memora-cli/README.md
@@ -0,0 +1,79 @@
+# Memora CLI
+
+Modern command-line interface for the Memora Temporal Semantic Memory System.
+
+## Installation
+
+```bash
+pip install memora-cli
+```
+
+## Configuration
+
+Set the API endpoint URL (defaults to `http://localhost:8080`):
+
+```bash
+export MEMORA_API_URL="http://localhost:8080"
+```
+
+## Commands
+
+### Search Memories
+
+```bash
+memora search alice "What did she say about AI?"
+memora search alice "hiking activities" --type world --max-tokens 8000
+memora search alice "recent events" --budget 150 --trace
+```
+
+### Think (Generate Answers)
+
+```bash
+memora think alice "What do you think about machine learning?"
+```
+
+### Store Memories
+
+Store a single memory:
+
+```bash
+memora put alice "Alice loves machine learning and AI"
+memora put alice "Today we discussed neural networks" --context "team meeting"
+
+# Async mode - returns immediately, processes in background
+memora put alice "Important note" --async
+```
+
+### Import Files
+
+Import memories from local files (.txt and .md):
+
+```bash
+# Import a single file
+memora put-files alice meeting-notes.txt
+
+# Import all files from a directory
+memora put-files alice ./documents/
+
+# Async mode - queue files for background processing
+memora put-files alice ./documents/ --async
+```
+
+### List Agents
+
+```bash
+memora agents
+```
+
+## Features
+
+- Beautiful TUI with Rich formatting (panels, tables, syntax highlighting)
+- Color-coded fact types (cyan=world, magenta=agent, yellow=opinion)
+- Progress bars and spinners for async operations
+- Tree views for file hierarchies
+- HTTP client (no direct database access needed)
+
+## Requirements
+
+- Python >= 3.11
+- Memora API server running
diff --git a/memora-cli/memora_cli/__init__.py b/memora-cli/memora_cli/__init__.py
new file mode 100644
index 00000000..fcc91240
--- /dev/null
+++ b/memora-cli/memora_cli/__init__.py
@@ -0,0 +1,5 @@
+"""
+Memora CLI - Modern command-line interface for the Memora Memory System.
+"""
+
+__version__ = "0.1.0"
diff --git a/memora-cli/memora_cli/main.py b/memora-cli/memora_cli/main.py
new file mode 100644
index 00000000..08dac3d2
--- /dev/null
+++ b/memora-cli/memora_cli/main.py
@@ -0,0 +1,565 @@
+"""
+Memora CLI - HTTP client for Memora API.
+"""
+
+import os
+from pathlib import Path
+from typing import Optional, List
+from datetime import datetime
+
+import typer
+import httpx
+from rich.console import Console
+from rich.table import Table
+from rich.panel import Panel
+from rich.progress import Progress, SpinnerColumn, TextColumn, BarColumn, TaskProgressColumn
+from rich.markdown import Markdown
+from rich import box
+from rich.tree import Tree
+
+app = typer.Typer(
+    name="memora",
+    help="Modern CLI for Memora - Temporal Semantic Memory System",
+    add_completion=False,
+)
+
+console = Console()
+
+
+def get_api_url():
+    """Get API URL from environment variable."""
+    api_url = os.getenv("MEMORA_API_URL", "http://localhost:8080")
+    return api_url.rstrip("/")
+
+
+def make_api_request(
+    method: str,
+    endpoint: str,
+    json_data: Optional[dict] = None,
+    timeout: float = 60.0,
+) -> dict:
+    """
+    Make an API request with proper error handling.
+
+    Args:
+        method: HTTP method (GET, POST, etc.)
+        endpoint: API endpoint path (e.g., "/api/search")
+        json_data: Optional JSON payload for POST requests
+        timeout: Request timeout in seconds
+
+    Returns:
+        Response data as dict
+
+    Raises:
+        typer.Exit on any error
+    """
+    api_url = get_api_url()
+    full_url = f"{api_url}{endpoint}"
+
+    try:
+        with httpx.Client(timeout=timeout) as client:
+            if method.upper() == "GET":
+                response = client.get(full_url)
+            elif method.upper() == "POST":
+                response = client.post(full_url, json=json_data)
+            else:
+                console.print(f"[red]Error: Unsupported HTTP method: {method}[/red]")
+                raise typer.Exit(1)
+
+            # Check HTTP status
+            response.raise_for_status()
+
+            # Parse response
+            data = response.json()
+
+            # Check for success field in response (if present)
+            if "success" in data and not data["success"]:
+                error_msg = data.get("message", "Unknown error")
+                console.print(f"[red]API Error: {error_msg}[/red]")
+                if "detail" in data:
+                    console.print(f"[yellow]Details: {data['detail']}[/yellow]")
+                raise typer.Exit(1)
+
+            return data
+
+    except httpx.HTTPStatusError as e:
+        console.print(f"[red]HTTP Error {e.response.status_code}[/red]")
+        try:
+            error_data = e.response.json()
+            if "detail" in error_data:
+                console.print(f"[red]Error: {error_data['detail']}[/red]")
+            else:
+                console.print(f"[red]Error: {error_data}[/red]")
+        except Exception:
+            console.print(f"[red]Error: {e.response.text}[/red]")
+        console.print(f"[yellow]Make sure the API is running at {api_url}[/yellow]")
+        raise typer.Exit(1)
+    except httpx.ConnectError as e:
+        console.print(f"[red]Connection Error: Failed to connect to API at {api_url}[/red]")
+        console.print(f"[yellow]Make sure the API server is running[/yellow]")
+        raise typer.Exit(1)
+    except httpx.TimeoutException:
+        console.print(f"[red]Timeout Error: Request took too long[/red]")
+        console.print(f"[yellow]Try increasing the timeout or check the API server[/yellow]")
+        raise typer.Exit(1)
+    except Exception as e:
+        console.print(f"[red]Unexpected Error: {e}[/red]")
+        raise typer.Exit(1)
+
+
+@app.command()
+def search(
+    agent_id: str = typer.Argument(..., help="Agent ID to search for"),
+    query: str = typer.Argument(..., help="Search query"),
+    fact_type: List[str] = typer.Option(
+        ["world", "agent", "opinion"],
+        "--type",
+        "-t",
+        help="Fact types to search (world/agent/opinion)",
+    ),
+    thinking_budget: int = typer.Option(
+        100, "--budget", "-b", help="Thinking budget for search"
+    ),
+    max_tokens: int = typer.Option(
+        4096, "--max-tokens", help="Maximum tokens for search results"
+    ),
+    trace: bool = typer.Option(False, "--trace", help="Show trace information"),
+):
+    """
+    Search memory using semantic similarity.
+
+    Example:
+        memora search alice "What did she say about AI?"
+    """
+    with console.status(f"[bold blue]Searching memories for {agent_id}...", spinner="dots"):
+        data = make_api_request(
+            method="POST",
+            endpoint="/api/search",
+            json_data={
+                "query": query,
+                "fact_type": list(fact_type),
+                "agent_id": agent_id,
+                "thinking_budget": thinking_budget,
+                "max_tokens": max_tokens,
+                "trace": trace,
+            },
+            timeout=60.0,
+        )
+
+    results = data.get("results", [])
+    trace_data = data.get("trace")
+
+    # Display results
+    if not results:
+        console.print("[yellow]No results found.[/yellow]")
+        return
+
+    console.print(f"\n[bold green]Found {len(results)} results:[/bold green]\n")
+
+    for i, result in enumerate(results, 1):
+        # Create a panel for each result
+        score = result.get("score", 0.0)
+        text = result.get("text", "")
+        fact_type_val = result.get("fact_type", "unknown")
+        context = result.get("context", "")
+        date = result.get("date", "")
+
+        # Color code based on fact type
+        type_colors = {
+            "world": "cyan",
+            "agent": "magenta",
+            "opinion": "yellow"
+        }
+        color = type_colors.get(fact_type_val, "white")
+
+        # Build info line
+        info_parts = [f"[{color}]{fact_type_val.upper()}[/{color}]"]
+        if context:
+            info_parts.append(f"Context: {context}")
+        if date:
+            info_parts.append(f"Date: {date}")
+        info_parts.append(f"Score: {score:.3f}")
+
+        info_line = " | ".join(info_parts)
+
+        panel = Panel(
+            f"{text}\n\n[dim]{info_line}[/dim]",
+            title=f"[bold]Result {i}[/bold]",
+            border_style=color,
+            box=box.ROUNDED,
+        )
+        console.print(panel)
+
+    # Show trace if requested
+    if trace and trace_data:
+        console.print("\n[bold blue]Trace Information:[/bold blue]")
+
+        trace_table = Table(show_header=True, box=box.SIMPLE)
+        trace_table.add_column("Metric", style="cyan")
+        trace_table.add_column("Value", style="green")
+
+        if "search_time_seconds" in trace_data:
+            trace_table.add_row("Search Time", f"{trace_data['search_time_seconds']:.3f}s")
+        if "total_activated" in trace_data:
+            trace_table.add_row("Total Activated", str(trace_data["total_activated"]))
+        if "results_returned" in trace_data:
+            trace_table.add_row("Results Returned", str(trace_data["results_returned"]))
+
+        console.print(trace_table)
+
+
+@app.command()
+def think(
+    agent_id: str = typer.Argument(..., help="Agent ID"),
+    query: str = typer.Argument(..., help="Question to think about"),
+    thinking_budget: int = typer.Option(
+        50, "--budget", "-b", help="Thinking budget"
+    ),
+):
+    """
+    Think and generate an answer using agent identity and memories.
+
+    Example:
+        memora think alice "What do you think about machine learning?"
+    """
+    with console.status(f"[bold blue]Thinking...", spinner="dots"):
+        result = make_api_request(
+            method="POST",
+            endpoint="/api/think",
+            json_data={
+                "query": query,
+                "agent_id": agent_id,
+                "thinking_budget": thinking_budget,
+            },
+            timeout=60.0,
+        )
+
+    # Display answer
+    console.print(Panel(
+        Markdown(result["text"]),
+        title=f"[bold cyan]Answer for {agent_id}[/bold cyan]",
+        border_style="cyan",
+        box=box.DOUBLE,
+    ))
+
+    # Display what the answer was based on
+    based_on = result.get("based_on", {})
+    if based_on:
+        console.print("\n[bold blue]Based on:[/bold blue]\n")
+
+        for fact_type, facts in based_on.items():
+            if facts:
+                type_colors = {
+                    "world": "cyan",
+                    "agent": "magenta",
+                    "opinion": "yellow"
+                }
+                color = type_colors.get(fact_type, "white")
+
+                table = Table(
+                    title=f"[{color}]{fact_type.upper()}[/{color}]",
+                    show_header=True,
+                    box=box.ROUNDED,
+                    border_style=color,
+                )
+                table.add_column("Text", style="white", width=80)
+                table.add_column("Score", justify="right", style="green", width=10)
+
+                for fact in facts[:5]:  # Show top 5
+                    text = fact.get("text", "")
+                    score = fact.get("score", 0.0)
+                    table.add_row(text, f"{score:.3f}")
+
+                console.print(table)
+
+    # Display new opinions formed
+    new_opinions = result.get("new_opinions", [])
+    if new_opinions:
+        console.print("\n[bold yellow]New Opinions Formed:[/bold yellow]\n")
+        for opinion in new_opinions:
+            console.print(Panel(
+                f"{opinion['text']}\n\n[dim]Confidence: {opinion['confidence']:.2f}[/dim]",
+                border_style="yellow",
+                box=box.ROUNDED,
+            ))
+
+
+@app.command()
+def put(
+    agent_id: str = typer.Argument(..., help="Agent ID"),
+    content: str = typer.Argument(..., help="Memory content to store"),
+    document_id: Optional[str] = typer.Option(
+        None, "--doc-id", "-d", help="Document ID (auto-generated if not provided)"
+    ),
+    context: Optional[str] = typer.Option(
+        None, "--context", "-c", help="Context for the memory"
+    ),
+    use_async: bool = typer.Option(
+        False, "--async", help="Use async batch put (returns immediately, processes in background)"
+    ),
+):
+    """
+    Store a memory from text input.
+
+    Example:
+        memora put alice "Alice loves machine learning and AI"
+        memora put alice "Today we discussed neural networks" --context "team meeting"
+        memora put alice "Important note" --async
+    """
+    # Generate document_id if not provided
+    if not document_id:
+        document_id = f"cli_put_{datetime.now().strftime('%Y%m%d_%H%M%S')}"
+
+    # Prepare content
+    item = {"content": content}
+    if context:
+        item["context"] = context
+
+    # Choose endpoint based on async flag
+    endpoint = "/api/memories/batch_async" if use_async else "/api/memories/batch"
+    status_msg = "Queueing memory" if use_async else "Storing memory"
+
+    with Progress(
+        SpinnerColumn(),
+        TextColumn("[progress.description]{task.description}"),
+        BarColumn(),
+        TaskProgressColumn(),
+        console=console,
+    ) as progress:
+        task = progress.add_task(f"[cyan]{status_msg} for {agent_id}...", total=None)
+
+        result = make_api_request(
+            method="POST",
+            endpoint=endpoint,
+            json_data={
+                "agent_id": agent_id,
+                "items": [item],
+                "document_id": document_id,
+            },
+            timeout=120.0,
+        )
+        progress.update(task, completed=True)
+
+    # Check if the result indicates success
+    if not result.get("success", False):
+        console.print(Panel(
+            f"[red]✗[/red] Failed to store memory\n"
+            f"[dim]Error:[/dim] {result.get('message', 'Unknown error')}",
+            title="[bold red]Storage Failed[/bold red]",
+            border_style="red",
+            box=box.ROUNDED,
+        ))
+        raise typer.Exit(1)
+
+    # Display result based on async vs sync
+    if use_async and result.get("queued", False):
+        console.print(Panel(
+            f"[green]✓[/green] Memory queued for background processing\n"
+            f"[dim]Agent ID:[/dim] {agent_id}\n"
+            f"[dim]Document ID:[/dim] {document_id}\n"
+            f"[dim]Content length:[/dim] {len(content)} characters\n"
+            f"[dim]Items queued:[/dim] {result.get('items_count', 1)}\n"
+            f"[yellow]Processing in background...[/yellow]",
+            title="[bold green]Memory Queued[/bold green]",
+            border_style="green",
+            box=box.ROUNDED,
+        ))
+    else:
+        console.print(Panel(
+            f"[green]✓[/green] Successfully stored memory\n"
+            f"[dim]Agent ID:[/dim] {agent_id}\n"
+            f"[dim]Document ID:[/dim] {document_id}\n"
+            f"[dim]Content length:[/dim] {len(content)} characters\n"
+            f"[dim]Items processed:[/dim] {result.get('items_count', 1)}",
+            title="[bold green]Memory Stored[/bold green]",
+            border_style="green",
+            box=box.ROUNDED,
+        ))
+
+
+@app.command(name="put-files")
+def put_files(
+    agent_id: str = typer.Argument(..., help="Agent ID"),
+    path: str = typer.Argument(..., help="File or directory path"),
+    recursive: bool = typer.Option(
+        True, "--recursive/--no-recursive", "-r", help="Search directories recursively"
+    ),
+    use_async: bool = typer.Option(
+        False, "--async", help="Use async batch put (returns immediately, processes in background)"
+    ),
+):
+    """
+    Store memories from local files (.txt and .md only).
+    Each file becomes a separate document with the filename as doc_id.
+
+    Example:
+        memora put-files alice ./documents/
+        memora put-files alice meeting-notes.txt
+        memora put-files alice ./documents/ --async
+    """
+    path_obj = Path(path)
+
+    if not path_obj.exists():
+        console.print(f"[red]Error: Path '{path}' does not exist[/red]")
+        raise typer.Exit(1)
+
+    # Collect files to process
+    files_to_process = []
+
+    if path_obj.is_file():
+        if path_obj.suffix.lower() in ['.txt', '.md']:
+            files_to_process.append(path_obj)
+        else:
+            console.print(f"[yellow]Warning: Skipping '{path}' - only .txt and .md files are supported[/yellow]")
+            raise typer.Exit(0)
+    else:
+        # Directory - find all .txt and .md files
+        pattern = "**/*" if recursive else "*"
+        for ext in ['.txt', '.md']:
+            files_to_process.extend(path_obj.glob(f"{pattern}{ext}"))
+
+        if not files_to_process:
+            console.print(f"[yellow]No .txt or .md files found in '{path}'[/yellow]")
+            raise typer.Exit(0)
+
+    # Display files to be processed
+    console.print(f"\n[bold]Found {len(files_to_process)} files to process:[/bold]\n")
+
+    tree = Tree(f"[bold cyan]{path}[/bold cyan]")
+    for file_path in sorted(files_to_process):
+        size = file_path.stat().st_size
+        size_str = f"{size:,} bytes" if size < 1024 else f"{size/1024:.1f} KB"
+        tree.add(f"{file_path.name} [dim]({size_str})[/dim]")
+    console.print(tree)
+    console.print()
+
+    # Process files
+    successful = 0
+    failed = 0
+    queued = 0
+
+    # Choose endpoint based on async flag
+    endpoint = "/api/memories/batch_async" if use_async else "/api/memories/batch"
+    status_msg = "Queueing files" if use_async else "Processing files"
+
+    with Progress(
+        SpinnerColumn(),
+        TextColumn("[progress.description]{task.description}"),
+        BarColumn(),
+        TaskProgressColumn(),
+        console=console,
+    ) as progress:
+        main_task = progress.add_task(
+            f"[cyan]{status_msg} for {agent_id}...",
+            total=len(files_to_process)
+        )
+
+        for file_path in files_to_process:
+            try:
+                # Read file content
+                content = file_path.read_text(encoding='utf-8')
+
+                # Use filename (without extension) as document_id
+                doc_id = file_path.stem
+
+                # Prepare content
+                item = {
+                    "content": content,
+                    "context": f"File: {file_path.name}"
+                }
+
+                # Store memory via API
+                result = make_api_request(
+                    method="POST",
+                    endpoint=endpoint,
+                    json_data={
+                        "agent_id": agent_id,
+                        "items": [item],
+                        "document_id": doc_id,
+                    },
+                    timeout=120.0,
+                )
+
+                # Check if the result indicates success
+                if not result.get("success", False):
+                    raise Exception(result.get("message", "Unknown error"))
+
+                if use_async and result.get("queued", False):
+                    queued += 1
+                else:
+                    successful += 1
+                progress.update(main_task, advance=1)
+
+            except typer.Exit:
+                # Re-raise typer.Exit to stop execution
+                raise
+            except Exception as e:
+                console.print(f"[red]Failed to process {file_path.name}: {str(e)}[/red]")
+                failed += 1
+                progress.update(main_task, advance=1)
+
+    # Summary
+    console.print()
+    if use_async and queued > 0:
+        console.print(Panel(
+            f"[green]✓[/green] Successfully queued {queued} file(s) for background processing\n"
+            f"[red]✗[/red] Failed: {failed}\n"
+            f"[dim]Agent ID:[/dim] {agent_id}\n"
+            f"[yellow]Processing in background...[/yellow]",
+            title="[bold green]Files Queued[/bold green]",
+            border_style="green" if failed == 0 else "yellow",
+            box=box.ROUNDED,
+        ))
+    elif successful > 0:
+        console.print(Panel(
+            f"[green]✓[/green] Successfully processed {successful} file(s)\n"
+            f"[red]✗[/red] Failed: {failed}\n"
+            f"[dim]Agent ID:[/dim] {agent_id}",
+            title="[bold green]Files Processed[/bold green]",
+            border_style="green" if failed == 0 else "yellow",
+            box=box.ROUNDED,
+        ))
+    else:
+        console.print("[red]No files were successfully processed[/red]")
+
+
+@app.command()
+def agents():
+    """
+    List all agents in the memory system.
+
+    Example:
+        memora agents
+    """
+    with console.status("[bold blue]Fetching agents...", spinner="dots"):
+        data = make_api_request(
+            method="GET",
+            endpoint="/api/agents",
+            timeout=30.0,
+        )
+
+    agent_list = data.get("agents", [])
+
+    if not agent_list:
+        console.print("[yellow]No agents found in the system.[/yellow]")
+        return
+
+    console.print(f"\n[bold green]Found {len(agent_list)} agent(s):[/bold green]\n")
+
+    table = Table(show_header=True, box=box.ROUNDED, border_style="cyan")
+    table.add_column("#", style="dim", width=6)
+    table.add_column("Agent ID", style="cyan")
+
+    for i, agent in enumerate(agent_list, 1):
+        table.add_row(str(i), agent)
+
+    console.print(table)
+
+
+def main():
+    """Main entry point for the CLI."""
+    app()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/memora-cli/pyproject.toml b/memora-cli/pyproject.toml
new file mode 100644
index 00000000..09020b64
--- /dev/null
+++ b/memora-cli/pyproject.toml
@@ -0,0 +1,21 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "memora-cli"
+version = "0.1.0"
+description = "Modern CLI for Memora - Temporal Semantic Memory System"
+readme = "README.md"
+requires-python = ">=3.11"
+dependencies = [
+    "rich>=13.0.0",
+    "typer>=0.20.0",
+    "httpx>=0.27.0",
+]
+
+[project.scripts]
+memora = "memora_cli.main:main"
+
+[tool.hatch.build.targets.wheel]
+packages = ["memora_cli"]
diff --git a/memora/memora/api.py b/memora/memora/api.py
index 462dd78d..48a9908f 100644
--- a/memora/memora/api.py
+++ b/memora/memora/api.py
@@ -21,7 +21,7 @@ from memora import TemporalSemanticMemory
 class SearchRequest(BaseModel):
     """Request model for search endpoint."""
     query: str
-    fact_type: List[str]  # List of fact types to search
+    fact_type: Optional[List[str]] = None  # List of fact types to search (defaults to all if not specified)
     agent_id: str = "default"
     thinking_budget: int = 100
     max_tokens: int = 4096
@@ -274,7 +274,6 @@ class ListDocumentsResponse(BaseModel):
                         "id": "session_1",
                         "agent_id": "user123",
                         "content_hash": "abc123",
-                        "metadata": {"source": "conversation"},
                         "created_at": "2024-01-15T10:30:00Z",
                         "updated_at": "2024-01-15T10:30:00Z",
                         "text_length": 5420,
@@ -294,7 +293,6 @@ class DocumentResponse(BaseModel):
     agent_id: str
     original_text: str
     content_hash: Optional[str]
-    metadata: Dict[str, Any]
     created_at: str
     updated_at: str
     memory_unit_count: int
@@ -306,7 +304,6 @@ class DocumentResponse(BaseModel):
                 "agent_id": "user123",
                 "original_text": "Full document text here...",
                 "content_hash": "abc123",
-                "metadata": {"source": "conversation"},
                 "created_at": "2024-01-15T10:30:00Z",
                 "updated_at": "2024-01-15T10:30:00Z",
                 "memory_unit_count": 15
@@ -380,15 +377,6 @@ The system uses:
 def _register_routes(app: FastAPI):
     """Register all API routes on the given app instance."""
 
-    @app.get("/", include_in_schema=False)
-    async def index():
-        """Root endpoint - directs to control plane."""
-        return {
-            "message": "Memory Control Plane API",
-            "docs": "/docs",
-            "control_plane": "The web UI has moved to the Next.js control plane. Please use the control-plane directory."
-        }
-
 
     @app.get(
         "/api/graph",
@@ -472,18 +460,16 @@ def _register_routes(app: FastAPI):
             # Validate fact_type(s)
             valid_fact_types = ["world", "agent", "opinion"]
 
+            # Default to all fact types if not specified
             if not request.fact_type:
-                raise HTTPException(
-                    status_code=400,
-                    detail="fact_type must be a non-empty list"
-                )
-
-            for ft in request.fact_type:
-                if ft not in valid_fact_types:
-                    raise HTTPException(
-                        status_code=400,
-                        detail=f"Invalid fact_type '{ft}'. Must be one of: {', '.join(valid_fact_types)}"
-                    )
+                request.fact_type = valid_fact_types
+            else:
+                for ft in request.fact_type:
+                    if ft not in valid_fact_types:
+                        raise HTTPException(
+                            status_code=400,
+                            detail=f"Invalid fact_type '{ft}'. Must be one of: {', '.join(valid_fact_types)}"
+                        )
 
             # Parse question_date if provided
             question_date = None
@@ -508,11 +494,22 @@ def _register_routes(app: FastAPI):
                 question_date=question_date
             )
 
+            # Filter results to only include specific fields
+            filtered_results = [
+                {
+                    "id": result.get("id"),
+                    "text": result.get("text"),
+                    "context": result.get("context"),
+                    "event_date": result.get("event_date")
+                }
+                for result in results
+            ]
+
             # Convert trace to dict
             trace_dict = trace.to_dict() if trace else None
 
             return SearchResponse(
-                results=results,
+                results=filtered_results,
                 trace=trace_dict
             )
         except HTTPException:
@@ -672,7 +669,7 @@ def _register_routes(app: FastAPI):
         description="List documents with pagination and optional search. Documents are the source content from which memory units are extracted."
     )
     async def api_list_documents(
-        agent_id: Optional[str] = None,
+        agent_id: str,
         q: Optional[str] = None,
         limit: int = 100,
         offset: int = 0
@@ -681,7 +678,7 @@ def _register_routes(app: FastAPI):
         List documents for an agent with optional search.
 
         Args:
-            agent_id: Filter by agent ID
+            agent_id: Agent ID (required)
             q: Search query (searches document ID and metadata)
             limit: Maximum number of results (default: 100)
             offset: Offset for pagination (default: 0)
@@ -760,14 +757,6 @@ def _register_routes(app: FastAPI):
     )
     async def api_batch_put(request: BatchPutRequest):
         try:
-            # Validate agent_id - prevent writing to reserved agents
-            RESERVED_AGENT_IDS = {"locomo"}
-            if request.agent_id in RESERVED_AGENT_IDS:
-                raise HTTPException(
-                    status_code=403,
-                    detail=f"Cannot write to reserved agent_id '{request.agent_id}'. Reserved agents: {', '.join(RESERVED_AGENT_IDS)}"
-                )
-
             # Prepare contents for put_batch_async
             contents = []
             for item in request.items:
@@ -784,7 +773,7 @@ def _register_routes(app: FastAPI):
                 contents=contents,
                 document_id=request.document_id
             )
-            logging.info(f"Batch put result: {result}")
+
 
             return BatchPutResponse(
                 success=True,
@@ -830,14 +819,6 @@ def _register_routes(app: FastAPI):
     )
     async def api_batch_put_async(request: BatchPutRequest):
         try:
-            # Validate agent_id - prevent writing to reserved agents
-            RESERVED_AGENT_IDS = {"locomo"}
-            if request.agent_id in RESERVED_AGENT_IDS:
-                raise HTTPException(
-                    status_code=403,
-                    detail=f"Cannot write to reserved agent_id '{request.agent_id}'. Reserved agents: {', '.join(RESERVED_AGENT_IDS)}"
-                )
-
             # Prepare contents for put_batch_async
             contents = []
             for item in request.items:
diff --git a/memora/memora/operations/think_operations.py b/memora/memora/operations/think_operations.py
index 755f0f64..37fc3c77 100644
--- a/memora/memora/operations/think_operations.py
+++ b/memora/memora/operations/think_operations.py
@@ -57,11 +57,15 @@ class ThinkOperationsMixin:
             fact_type=['agent', 'world', 'opinion']
         )
 
+        logger.info(f"[THINK] Search returned {len(all_results)} results")
+
         # Split results by fact type for structured response
         agent_results = [r for r in all_results if r.get('fact_type') == 'agent']
         world_results = [r for r in all_results if r.get('fact_type') == 'world']
         opinion_results = [r for r in all_results if r.get('fact_type') == 'opinion']
 
+        logger.info(f"[THINK] Split results - agent: {len(agent_results)}, world: {len(world_results)}, opinion: {len(opinion_results)}")
+
         # Step 4: Format facts for LLM with full details as JSON
         import json
 
@@ -99,6 +103,8 @@ class ThinkOperationsMixin:
         world_facts_text = format_facts(world_results)
         opinion_facts_text = format_facts(opinion_results)
 
+        logger.info(f"[THINK] Formatted facts - agent: {len(agent_facts_text)} chars, world: {len(world_facts_text)} chars, opinion: {len(opinion_facts_text)} chars")
+
         # Step 5: Call Groq to formulate answer
         prompt = f"""You are an AI assistant answering a question based on retrieved facts provided in JSON format.
 
@@ -123,6 +129,9 @@ Provide a helpful, accurate answer based on the facts above. Be consistent with
 
 If you form any new opinions while thinking about this question, state them clearly in your answer."""
 
+        logger.info(f"[THINK] Full prompt length: {len(prompt)} chars")
+        logger.debug(f"[THINK] Prompt preview (first 500 chars): {prompt[:500]}")
+
         answer_text = await self._llm_config.call(
             messages=[
                 {"role": "system", "content": "You are a helpful AI assistant. Always respond in plain text without markdown formatting. You can form and express opinions based on facts."},
diff --git a/memora/memora/temporal_semantic_memory.py b/memora/memora/temporal_semantic_memory.py
index bbbfecf5..1576c5a8 100644
--- a/memora/memora/temporal_semantic_memory.py
+++ b/memora/memora/temporal_semantic_memory.py
@@ -1281,6 +1281,7 @@ class TemporalSemanticMemory(
                     "text": data["text"],
                     "context": data.get("context", ""),
                     "event_date": data["event_date"],  # Keep as datetime for now
+                    "fact_type": data.get("fact_type"),  # Include fact type for filtering
                     "access_count": data.get("access_count", 0),
                     "semantic_similarity": semantic_sim,
                     "bm25_score": bm25_score,
@@ -1447,12 +1448,12 @@ class TemporalSemanticMemory(
         async with pool.acquire() as conn:
             doc = await conn.fetchrow(
                 """
-                SELECT d.id, d.agent_id, d.original_text, d.content_hash, d.metadata,
+                SELECT d.id, d.agent_id, d.original_text, d.content_hash,
                        d.created_at, d.updated_at, COUNT(mu.id) as unit_count
                 FROM documents d
                 LEFT JOIN memory_units mu ON mu.document_id = d.id
                 WHERE d.id = $1 AND d.agent_id = $2
-                GROUP BY d.id, d.agent_id, d.original_text, d.content_hash, d.metadata, d.created_at, d.updated_at
+                GROUP BY d.id, d.agent_id, d.original_text, d.content_hash, d.created_at, d.updated_at
                 """,
                 document_id, agent_id
             )
@@ -1460,13 +1461,11 @@ class TemporalSemanticMemory(
             if not doc:
                 return None
 
-            import json
             return {
                 "id": doc["id"],
                 "agent_id": doc["agent_id"],
                 "original_text": doc["original_text"],
                 "content_hash": doc["content_hash"],
-                "metadata": json.loads(doc["metadata"]) if doc["metadata"] else {},
                 "unit_count": doc["unit_count"],
                 "created_at": doc["created_at"],
                 "updated_at": doc["updated_at"]
@@ -1871,7 +1870,7 @@ class TemporalSemanticMemory(
 
     async def list_documents(
         self,
-        agent_id: Optional[str] = None,
+        agent_id: str,
         search_query: Optional[str] = None,
         limit: int = 100,
         offset: int = 0
@@ -1880,8 +1879,8 @@ class TemporalSemanticMemory(
         List documents with optional search and pagination.
 
         Args:
-            agent_id: Filter by agent ID
-            search_query: Search in metadata (JSON text search)
+            agent_id: Agent ID (required)
+            search_query: Search in document ID
             limit: Maximum number of results
             offset: Offset for pagination
 
@@ -1895,15 +1894,14 @@ class TemporalSemanticMemory(
             query_params = []
             param_count = 0
 
-            if agent_id:
-                param_count += 1
-                query_conditions.append(f"agent_id = ${param_count}")
-                query_params.append(agent_id)
+            param_count += 1
+            query_conditions.append(f"agent_id = ${param_count}")
+            query_params.append(agent_id)
 
             if search_query:
-                # Search in document ID and metadata (as text)
+                # Search in document ID
                 param_count += 1
-                query_conditions.append(f"(id ILIKE ${param_count} OR metadata::text ILIKE ${param_count})")
+                query_conditions.append(f"id ILIKE ${param_count}")
                 query_params.append(f"%{search_query}%")
 
             where_clause = "WHERE " + " AND ".join(query_conditions) if query_conditions else ""
@@ -1931,7 +1929,6 @@ class TemporalSemanticMemory(
                     id,
                     agent_id,
                     content_hash,
-                    metadata,
                     created_at,
                     updated_at,
                     LENGTH(original_text) as text_length
@@ -1979,7 +1976,6 @@ class TemporalSemanticMemory(
                     "id": doc_id,
                     "agent_id": agent_id_val,
                     "content_hash": row['content_hash'],
-                    "metadata": row['metadata'] if row['metadata'] else {},
                     "created_at": row['created_at'].isoformat() if row['created_at'] else "",
                     "updated_at": row['updated_at'].isoformat() if row['updated_at'] else "",
                     "text_length": row['text_length'] or 0,
@@ -2016,7 +2012,6 @@ class TemporalSemanticMemory(
                     agent_id,
                     original_text,
                     content_hash,
-                    metadata,
                     created_at,
                     updated_at
                 FROM documents
@@ -2038,7 +2033,6 @@ class TemporalSemanticMemory(
                 "agent_id": doc['agent_id'],
                 "original_text": doc['original_text'],
                 "content_hash": doc['content_hash'],
-                "metadata": doc['metadata'] if doc['metadata'] else {},
                 "created_at": doc['created_at'].isoformat() if doc['created_at'] else "",
                 "updated_at": doc['updated_at'].isoformat() if doc['updated_at'] else "",
                 "memory_unit_count": unit_count_row['unit_count'] if unit_count_row else 0
diff --git a/memora/tests/test_performance_tuning.py b/memora/tests/test_performance_tuning.py
deleted file mode 100644
index ea3f8ad0..00000000
--- a/memora/tests/test_performance_tuning.py
+++ /dev/null
@@ -1,114 +0,0 @@
-"""
-Performance tuning test using real LoComo conversation.
-
-This test loads a long conversation (419 dialogues across 19 sessions),
-ingests it into memory, and runs searches to measure performance.
-"""
-import logging
-import json
-import pytest
-from datetime import datetime, timezone
-from pathlib import Path
-
-
-# Configure logging to show performance metrics
-logging.basicConfig(
-    level=logging.INFO,
-    format='%(asctime)s %(levelname)s:%(name)s: %(message)s'
-)
-
-
-@pytest.mark.asyncio
-@pytest.mark.timeout(300)  # 5 minute timeout for performance test
-async def test_batch_ingestion_single_call(memory):
-    """
-    Test ingesting entire conversation in ONE batch call.
-
-    This is the most efficient way - all sessions in one put_batch_async.
-    """
-    # Load conversation fixture
-    fixture_path = Path(__file__).parent / "fixtures" / "locomo_conversation_sample.json"
-    with open(fixture_path) as f:
-        conversation_data = json.load(f)
-
-    sample_id = conversation_data['sample_id']
-    logging.info(f"\n{'='*80}")
-    logging.info(f"BATCH INGESTION TEST: {sample_id}")
-    logging.info(f"{'='*80}")
-
-    agent_id = f"batch_test_{sample_id}_{datetime.now(timezone.utc).timestamp()}"
-
-    try:
-        # Parse all sessions into batch format
-        # LIMIT to first 5 sessions for faster iteration during perf tuning
-        MAX_SESSIONS = 5
-        logging.info(f"\nPreparing batch contents (limiting to {MAX_SESSIONS} sessions for perf tuning)...")
-        conversation = conversation_data['conversation']
-        batch_contents = []
-
-        for i in range(1, MAX_SESSIONS + 1):
-            session_key = f'session_{i}'
-            session_date_key = f'session_{i}_date_time'
-
-            if session_key not in conversation or not conversation[session_key]:
-                break
-
-            session_dialogues = conversation[session_key]
-            session_date = conversation.get(session_date_key, datetime.now(timezone.utc).isoformat())
-
-            # Combine dialogues
-            session_text = "\n".join([
-                f"{d['speaker']}: {d['text']}"
-                for d in session_dialogues
-            ])
-
-            # Parse date
-            try:
-                from dateutil import parser as date_parser
-                event_date = date_parser.isoparse(session_date)
-            except:
-                event_date = datetime.now(timezone.utc)
-
-            batch_contents.append({
-                'content': session_text,
-                'context': f'session_{i}',
-                'event_date': event_date
-            })
-
-        logging.info(f"Prepared {len(batch_contents)} sessions for batch ingestion")
-
-        # Single batch call
-        logging.info(f"\nIngesting all {len(batch_contents)} sessions in ONE batch call...")
-        result_ids = await memory.put_batch_async(
-            agent_id=agent_id,
-            contents=batch_contents,
-            document_id=f"{agent_id}_full_conversation"
-        )
-
-        total_units = sum(len(ids) for ids in result_ids)
-        logging.info(f"\n{'='*80}")
-        logging.info(f"BATCH INGESTION COMPLETE: {total_units} memory units created")
-        logging.info(f"{'='*80}")
-
-        # Run one sample search
-        logging.info(f"\nRunning sample search...")
-        question = conversation_data['qa'][0]['question']
-        logging.info(f"Question: {question}")
-
-        results, _ = await memory.search_async(
-            agent_id=agent_id,
-            query=question,
-            fact_type=["world"],
-            thinking_budget=100,
-            top_k=5,
-            enable_trace=False
-        )
-
-        logging.info(f"Found {len(results)} results")
-        if results:
-            logging.info(f"Top result: {results[0]['text'][:100]}...")
-
-    finally:
-        # Cleanup
-        logging.info("\nCleaning up...")
-        await memory.delete_agent(agent_id)
diff --git a/pyproject.toml b/pyproject.toml
index 2bad4e60..ea1fd490 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,5 +1,5 @@
 [tool.uv.workspace]
-members = ["memora", "benchmarks", "memora-dev"]
+members = ["memora", "benchmarks", "memora-dev", "memora-cli"]
 
 [tool.uv]
 dev-dependencies = []
diff --git a/scripts/start-control-plane.sh b/scripts/start-control-plane.sh
new file mode 100755
index 00000000..3d02de94
--- /dev/null
+++ b/scripts/start-control-plane.sh
@@ -0,0 +1,56 @@
+#!/bin/bash
+set -e
+
+cd "$(dirname "$0")/../control-plane"
+
+# Parse arguments
+PORT=3000
+
+while [[ $# -gt 0 ]]; do
+  case $1 in
+    --port|-p)
+      PORT="$2"
+      shift 2
+      ;;
+    --help|-h)
+      echo "Usage: $0 [options]"
+      echo ""
+      echo "Options:"
+      echo "  --port, -p PORT    Port to run on (default: 3000)"
+      echo "  --help, -h         Show this help message"
+      echo ""
+      echo "Example:"
+      echo "  $0 --port 3001"
+      exit 0
+      ;;
+    *)
+      echo "Unknown option: $1"
+      echo "Use --help for usage information"
+      exit 1
+      ;;
+  esac
+done
+
+# Check if .env.local exists
+if [ ! -f ".env.local" ]; then
+  echo "⚠️  Warning: .env.local not found"
+  echo "Creating from .env.local.example..."
+  if [ -f ".env.local.example" ]; then
+    cp .env.local.example .env.local
+    echo "✅ Created .env.local"
+    echo "📝 Please edit .env.local if you need to change the DATAPLANE_API_URL"
+    echo ""
+  else
+    echo "❌ Error: .env.local.example not found"
+    exit 1
+  fi
+fi
+
+echo "🚀 Starting Control Plane (Next.js dev server)..."
+echo "📄 Loading environment from .env.local"
+echo ""
+echo "Control plane will be available at: http://localhost:${PORT}"
+echo ""
+
+# Set the port and run dev server
+PORT=$PORT npm run dev
diff --git a/standalone/.dockerignore b/standalone/.dockerignore
new file mode 100644
index 00000000..905f41ac
--- /dev/null
+++ b/standalone/.dockerignore
@@ -0,0 +1,46 @@
+# Build artifacts
+**/*.pyc
+**/__pycache__/
+**/.pytest_cache/
+**/.venv/
+**/venv/
+**/*.egg-info/
+**/dist/
+**/build/
+
+# Node
+**/node_modules/
+**/.next/
+**/npm-debug.log
+**/.turbo/
+
+# Environment files
+.env
+.env.*
+!standalone/.env.standalone
+
+# Git
+.git/
+.gitignore
+.gitattributes
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+
+# Test and dev files
+**/tests/
+local-db/
+logs/
+
+# Documentation (except standalone README)
+README.md
+!standalone/README.md
+
+# Standalone files
+standalone/build-docker.sh
+standalone/.dockerignore
+standalone/.env.example
+standalone/docker-compose.yml
diff --git a/standalone/.env.example b/standalone/.env.example
new file mode 100644
index 00000000..64bc282b
--- /dev/null
+++ b/standalone/.env.example
@@ -0,0 +1,9 @@
+# Environment variables for docker-compose
+# Copy this file to .env and customize as needed
+
+# Optional: OpenAI API key
+# OPENAI_API_KEY=your-api-key-here
+
+# Optional: Custom embedding model
+# EMBEDDING_MODEL_NAME=sentence-transformers/all-MiniLM-L6-v2
+# EMBEDDING_DIM=384
diff --git a/standalone/.env.standalone b/standalone/.env.standalone
new file mode 100644
index 00000000..05bd6a1a
--- /dev/null
+++ b/standalone/.env.standalone
@@ -0,0 +1,15 @@
+# Standalone environment configuration
+DATABASE_URL=postgresql://postgres:postgres@localhost:5432/memora
+DATAPLANE_API_URL=http://localhost:8080
+
+# Embedding configuration
+EMBEDDING_MODEL_NAME=sentence-transformers/all-MiniLM-L6-v2
+EMBEDDING_DIM=384
+
+# LLM Provider (set to "none" to disable LLM features)
+LLM_PROVIDER=none
+
+# Optional: LLM API Keys
+# OPENAI_API_KEY=your-openai-key-here
+# ANTHROPIC_API_KEY=your-anthropic-key-here
+# GROQ_API_KEY=your-groq-key-here
diff --git a/standalone/.gitignore b/standalone/.gitignore
new file mode 100644
index 00000000..844a5129
--- /dev/null
+++ b/standalone/.gitignore
@@ -0,0 +1,5 @@
+# Environment files
+.env
+
+# Docker volumes
+*.log
diff --git a/standalone/Dockerfile b/standalone/Dockerfile
new file mode 100644
index 00000000..e07f184b
--- /dev/null
+++ b/standalone/Dockerfile
@@ -0,0 +1,87 @@
+FROM node:20-alpine AS control-plane-builder
+
+# Build control plane
+WORKDIR /app/control-plane
+COPY control-plane/package*.json ./
+RUN npm ci
+
+COPY control-plane/ ./
+# Set env to skip font optimization during build
+ENV NEXT_TELEMETRY_DISABLED=1
+RUN npm run build || (echo "Build failed, retrying..." && npm run build)
+
+# Python source stage - just copy files, don't build venv yet
+FROM python:3.12-slim AS dataplane-source
+
+WORKDIR /build
+COPY pyproject.toml uv.lock ./
+COPY memora/ ./memora/
+COPY benchmarks/ ./benchmarks/
+COPY memora-dev/ ./memora-dev/
+COPY memora-cli/ ./memora-cli/
+
+# Final runtime image
+FROM python:3.12-slim
+
+# Install system dependencies and PostgreSQL
+RUN apt-get update && apt-get install -y \
+    gnupg \
+    lsb-release \
+    wget \
+    curl \
+    ca-certificates \
+    && mkdir -p /etc/apt/keyrings \
+    && wget --quiet -O - https://www.postgresql.org/media/keys/ACCC4CF8.asc | gpg --dearmor -o /etc/apt/keyrings/pgdg.gpg \
+    && echo "deb [signed-by=/etc/apt/keyrings/pgdg.gpg] http://apt.postgresql.org/pub/repos/apt $(lsb_release -cs)-pgdg main" > /etc/apt/sources.list.d/pgdg.list \
+    && apt-get update && apt-get install -y \
+    postgresql-15 \
+    postgresql-15-pgvector \
+    postgresql-contrib-15 \
+    nodejs \
+    npm \
+    supervisor \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install uv
+RUN pip install uv
+
+# Create app directory
+WORKDIR /app
+
+# Copy dataplane source from builder
+COPY --from=dataplane-source /build /app
+
+# Build venv in the final stage to ensure compatibility
+RUN cd /app && uv sync --frozen
+
+# Copy control plane from builder
+COPY --from=control-plane-builder /app/control-plane/.next/standalone /app/control-plane
+COPY --from=control-plane-builder /app/control-plane/.next/static /app/control-plane/.next/static
+
+# Copy standalone configuration
+COPY standalone/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
+COPY standalone/init.sh /app/init.sh
+COPY standalone/.env.standalone /app/.env
+
+RUN chmod +x /app/init.sh
+
+# PostgreSQL setup
+RUN mkdir -p /var/lib/postgresql/data && \
+    chown -R postgres:postgres /var/lib/postgresql && \
+    mkdir -p /var/run/postgresql && \
+    chown -R postgres:postgres /var/run/postgresql
+
+# Initialize PostgreSQL as postgres user
+USER postgres
+RUN /usr/lib/postgresql/15/bin/initdb -D /var/lib/postgresql/data
+
+USER root
+
+# Expose ports
+# 5432: PostgreSQL
+# 8080: Dataplane API
+# 3000: Control Plane
+EXPOSE 5432 8080 3000
+
+# Start supervisor
+CMD ["/app/init.sh"]
diff --git a/standalone/QUICKSTART.md b/standalone/QUICKSTART.md
new file mode 100644
index 00000000..cf1da00a
--- /dev/null
+++ b/standalone/QUICKSTART.md
@@ -0,0 +1,104 @@
+# Memora Standalone - Quick Start
+
+## What is this?
+
+A single Docker image containing everything you need to run Memora:
+- ✅ PostgreSQL database
+- ✅ Dataplane API (FastAPI backend)
+- ✅ Control Plane (Next.js web UI)
+
+## Fastest Start (Docker Compose)
+
+```bash
+cd standalone
+docker-compose up -d
+```
+
+Access the UI at: **http://localhost:3000**
+
+## Manual Docker Build & Run
+
+### Build the image:
+```bash
+./standalone/build-docker.sh
+```
+
+### Run with the helper script:
+```bash
+./standalone/run-docker.sh --persist
+```
+
+### Or run directly:
+```bash
+docker run -d \
+  --name memora \
+  -p 3000:3000 \
+  -p 8080:8080 \
+  -p 5432:5432 \
+  -v memora-data:/var/lib/postgresql/data \
+  memora-standalone:latest
+```
+
+## Access Points
+
+| Service | URL | Purpose |
+|---------|-----|---------|
+| **Control Plane** | http://localhost:3000 | Web UI |
+| **Dataplane API** | http://localhost:8080 | REST API |
+| **PostgreSQL** | localhost:5432 | Database |
+
+## View Logs
+
+```bash
+docker logs -f memora-standalone
+```
+
+## Stop & Remove
+
+```bash
+# Stop
+docker-compose down
+
+# Stop and remove data
+docker-compose down -v
+```
+
+## Environment Variables
+
+Set in `docker-compose.yml` or pass with `-e`:
+
+- `EMBEDDING_MODEL_NAME` - Sentence transformer model (default: sentence-transformers/all-MiniLM-L6-v2)
+- `EMBEDDING_DIM` - Embedding dimension (default: 384)
+- `OPENAI_API_KEY` - Optional OpenAI API key
+
+## Troubleshooting
+
+**Container won't start:**
+```bash
+docker logs memora-standalone
+```
+
+**Database issues:**
+```bash
+docker exec -it memora-standalone su - postgres -c "psql memora"
+```
+
+**Reset everything:**
+```bash
+docker-compose down -v
+docker-compose up -d
+```
+
+## Production Notes
+
+This standalone image is ideal for:
+- ✅ Development
+- ✅ Demos
+- ✅ Testing
+- ✅ Small deployments
+
+For production, consider:
+- Separate containers for each service
+- External PostgreSQL database
+- Kubernetes/Docker Swarm orchestration
+- Environment-specific configurations
diff --git a/standalone/README.md b/standalone/README.md
new file mode 100644
index 00000000..827eecf2
--- /dev/null
+++ b/standalone/README.md
@@ -0,0 +1,124 @@
+# Memora Standalone Docker Image
+
+This directory contains the configuration to build a standalone Docker image that includes all Memora components in a single container:
+
+- **PostgreSQL**: Database backend
+- **Dataplane**: FastAPI backend service
+- **Control Plane**: Next.js web interface
+
+## Quick Start with Docker Compose
+
+The easiest way to run the standalone image:
+
+```bash
+cd standalone
+docker-compose up -d
+```
+
+This will build and start all services with persistent data storage.
+
+To stop:
+```bash
+docker-compose down
+```
+
+To remove data and start fresh:
+```bash
+docker-compose down -v
+```
+
+## Building Manually
+
+```bash
+./standalone/build-docker.sh
+```
+
+With custom options:
+```bash
+./standalone/build-docker.sh --name my-memora --tag v1.0.0
+./standalone/build-docker.sh --registry docker.io/myuser --tag latest
+```
+
+## Running Manually
+
+Using the run script (recommended):
+```bash
+./standalone/run-docker.sh --persist
+```
+
+With custom ports:
+```bash
+./standalone/run-docker.sh --persist --port-control 3001 --port-api 8081
+```
+
+Direct docker run:
+```bash
+docker run -p 3000:3000 -p 8080:8080 memora-standalone:latest
+```
+
+With persistent data:
+```bash
+docker run -p 3000:3000 -p 8080:8080 \
+  -v memora-data:/var/lib/postgresql/data \
+  memora-standalone:latest
+```
+
+With custom environment variables:
+```bash
+docker run -p 3000:3000 -p 8080:8080 \
+  -e OPENAI_API_KEY=your-key \
+  -e EMBEDDING_MODEL_NAME=custom-model \
+  memora-standalone:latest
+```
+
+## Accessing Services
+
+Once running, services are available at:
+
+- **Control Plane**: http://localhost:3000
+- **Dataplane API**: http://localhost:8080
+- **PostgreSQL**: localhost:5432 (username: postgres, password: postgres, database: memora)
+
+## Architecture
+
+The container uses `supervisord` to manage three processes:
+1. PostgreSQL (started first)
+2. Dataplane API (started after PostgreSQL)
+3. Control Plane (started after dataplane)
+
+The `init.sh` script handles:
+- PostgreSQL initialization
+- Database creation
+- Running migrations
+- Starting all services via supervisord
+
+## Environment Variables
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `DATABASE_URL` | `postgresql://postgres:postgres@localhost:5432/memora` | PostgreSQL connection string |
+| `DATAPLANE_API_URL` | `http://localhost:8080` | Dataplane API URL for control plane |
+| `EMBEDDING_MODEL_NAME` | `sentence-transformers/all-MiniLM-L6-v2` | Sentence transformer model |
+| `EMBEDDING_DIM` | `384` | Embedding dimension |
+| `OPENAI_API_KEY` | - | Optional OpenAI API key |
+
+## Logs
+
+View logs from all services:
+```bash
+docker logs -f <container-id>
+```
+
+## Production Considerations
+
+This standalone image is designed for:
+- Development environments
+- Demos and testing
+- Small deployments
+
+For production use, consider:
+- Using separate containers for each service
+- External PostgreSQL database
+- Load balancing for the control plane
+- Persistent volume for PostgreSQL data
+- Environment-specific configurations
diff --git a/standalone/build-docker.sh b/standalone/build-docker.sh
new file mode 100755
index 00000000..a97d2cbf
--- /dev/null
+++ b/standalone/build-docker.sh
@@ -0,0 +1,87 @@
+#!/bin/bash
+set -e
+
+cd "$(dirname "$0")/.."
+
+# Default values
+IMAGE_NAME="memora-standalone"
+IMAGE_TAG="latest"
+REGISTRY=""
+
+# Parse command line arguments
+while [[ $# -gt 0 ]]; do
+    case $1 in
+        --name)
+            IMAGE_NAME="$2"
+            shift 2
+            ;;
+        --tag)
+            IMAGE_TAG="$2"
+            shift 2
+            ;;
+        --registry)
+            REGISTRY="$2"
+            shift 2
+            ;;
+        --help)
+            echo "Usage: $0 [OPTIONS]"
+            echo ""
+            echo "Options:"
+            echo "  --name NAME        Docker image name (default: memora-standalone)"
+            echo "  --tag TAG          Docker image tag (default: latest)"
+            echo "  --registry REG     Docker registry URL (optional)"
+            echo "  --help             Show this help message"
+            echo ""
+            echo "Example:"
+            echo "  $0 --name myapp --tag v1.0.0"
+            echo "  $0 --registry docker.io/myuser --name memora-standalone --tag v1.0.0"
+            exit 0
+            ;;
+        *)
+            echo "Unknown option: $1"
+            echo "Use --help for usage information"
+            exit 1
+            ;;
+    esac
+done
+
+# Construct full image name
+if [ -n "$REGISTRY" ]; then
+    FULL_IMAGE_NAME="${REGISTRY}/${IMAGE_NAME}:${IMAGE_TAG}"
+else
+    FULL_IMAGE_NAME="${IMAGE_NAME}:${IMAGE_TAG}"
+fi
+
+echo "Building Memora Standalone Docker image: ${FULL_IMAGE_NAME}"
+echo "============================================================="
+echo "This image includes:"
+echo "  - PostgreSQL database"
+echo "  - Dataplane API (FastAPI)"
+echo "  - Control Plane (Next.js)"
+echo ""
+
+# Build the Docker image
+docker build -f standalone/Dockerfile -t "${FULL_IMAGE_NAME}" .
+
+echo ""
+echo "Build completed successfully!"
+echo "Image: ${FULL_IMAGE_NAME}"
+echo ""
+echo "To run the container:"
+echo "  docker run -p 3000:3000 -p 8080:8080 ${FULL_IMAGE_NAME}"
+echo ""
+echo "Services will be available at:"
+echo "  - Control Plane: http://localhost:3000"
+echo "  - Dataplane API: http://localhost:8080"
+echo "  - PostgreSQL: localhost:5432"
+echo ""
+echo "For persistent data, mount a volume:"
+echo "  docker run -p 3000:3000 -p 8080:8080 \\"
+echo "    -v memora-data:/var/lib/postgresql/data \\"
+echo "    ${FULL_IMAGE_NAME}"
+echo ""
+if [ -n "$REGISTRY" ]; then
+    echo "To push to registry:"
+    echo "  docker push ${FULL_IMAGE_NAME}"
+    echo ""
+fi
diff --git a/standalone/docker-compose.yml b/standalone/docker-compose.yml
new file mode 100644
index 00000000..7ea0bf1b
--- /dev/null
+++ b/standalone/docker-compose.yml
@@ -0,0 +1,21 @@
+version: '3.8'
+
+services:
+  memora-standalone:
+    build:
+      context: ..
+      dockerfile: standalone/Dockerfile
+    ports:
+      - "3000:3000"  # Control Plane
+      - "8080:8080"  # Dataplane API
+      - "5432:5432"  # PostgreSQL
+    environment:
+      - EMBEDDING_MODEL_NAME=sentence-transformers/all-MiniLM-L6-v2
+      - EMBEDDING_DIM=384
+      # - OPENAI_API_KEY=${OPENAI_API_KEY}  # Uncomment if needed
+    volumes:
+      - memora-data:/var/lib/postgresql/data
+    restart: unless-stopped
+
+volumes:
+  memora-data:
diff --git a/standalone/init.sh b/standalone/init.sh
new file mode 100755
index 00000000..10609722
--- /dev/null
+++ b/standalone/init.sh
@@ -0,0 +1,55 @@
+#!/bin/bash
+set -e
+
+echo "🚀 Starting Memora Standalone Container..."
+echo "==========================================="
+
+# Start PostgreSQL temporarily for initialization
+echo "📦 Starting PostgreSQL for initialization..."
+su - postgres -c "/usr/lib/postgresql/15/bin/pg_ctl -D /var/lib/postgresql/data -l /tmp/postgresql-init.log start"
+
+# Wait for PostgreSQL to be ready
+echo "⏳ Waiting for PostgreSQL to be ready..."
+for i in {1..30}; do
+    if su - postgres -c "psql -lqt" &>/dev/null; then
+        echo "✅ PostgreSQL is ready"
+        break
+    fi
+    if [ $i -eq 30 ]; then
+        echo "❌ PostgreSQL failed to start"
+        cat /tmp/postgresql-init.log
+        exit 1
+    fi
+    sleep 1
+done
+
+# Create database if it doesn't exist
+echo "📊 Setting up database..."
+su - postgres -c "psql -tc \"SELECT 1 FROM pg_database WHERE datname = 'memora'\" | grep -q 1 || psql -c 'CREATE DATABASE memora;'"
+
+# Run migrations
+echo "🔄 Running database migrations..."
+cd /app/memora
+
+# Export environment variables
+set -a
+source /app/.env
+set +a
+
+/app/.venv/bin/python -m alembic upgrade head
+
+# Stop PostgreSQL so supervisord can start it cleanly
+echo "🔄 Stopping PostgreSQL to hand off to supervisord..."
+su - postgres -c "/usr/lib/postgresql/15/bin/pg_ctl -D /var/lib/postgresql/data stop -m fast"
+sleep 2
+
+echo "✅ Initialization complete"
+echo ""
+echo "Starting services via supervisord..."
+echo "  - PostgreSQL: localhost:5432"
+echo "  - Dataplane API: http://localhost:8080"
+echo "  - Control Plane: http://localhost:3000"
+echo ""
+
+# Start supervisor to manage all services
+exec /usr/bin/supervisord -c /etc/supervisor/conf.d/supervisord.conf
diff --git a/standalone/run-docker.sh b/standalone/run-docker.sh
new file mode 100755
index 00000000..ada21582
--- /dev/null
+++ b/standalone/run-docker.sh
@@ -0,0 +1,122 @@
+#!/bin/bash
+set -e
+
+# Default values
+IMAGE_NAME="memora-standalone:latest"
+CONTAINER_NAME="memora-standalone"
+PERSIST_DATA=false
+PORT_CONTROL=3000
+PORT_API=8080
+PORT_DB=5432
+
+# Parse command line arguments
+while [[ $# -gt 0 ]]; do
+    case $1 in
+        --image)
+            IMAGE_NAME="$2"
+            shift 2
+            ;;
+        --name)
+            CONTAINER_NAME="$2"
+            shift 2
+            ;;
+        --persist)
+            PERSIST_DATA=true
+            shift
+            ;;
+        --port-control)
+            PORT_CONTROL="$2"
+            shift 2
+            ;;
+        --port-api)
+            PORT_API="$2"
+            shift 2
+            ;;
+        --port-db)
+            PORT_DB="$2"
+            shift 2
+            ;;
+        --help)
+            echo "Usage: $0 [OPTIONS]"
+            echo ""
+            echo "Options:"
+            echo "  --image NAME           Docker image name (default: memora-standalone:latest)"
+            echo "  --name NAME            Container name (default: memora-standalone)"
+            echo "  --persist              Use persistent volume for data"
+            echo "  --port-control PORT    Control plane port (default: 3000)"
+            echo "  --port-api PORT        Dataplane API port (default: 8080)"
+            echo "  --port-db PORT         PostgreSQL port (default: 5432)"
+            echo "  --help                 Show this help message"
+            echo ""
+            echo "Example:"
+            echo "  $0 --persist --port-control 3001"
+            echo ""
+            echo "To stop the container:"
+            echo "  docker stop ${CONTAINER_NAME}"
+            echo ""
+            echo "To remove the container:"
+            echo "  docker rm ${CONTAINER_NAME}"
+            exit 0
+            ;;
+        *)
+            echo "Unknown option: $1"
+            echo "Use --help for usage information"
+            exit 1
+            ;;
+    esac
+done
+
+# Check if container already exists
+if docker ps -a --format '{{.Names}}' | grep -q "^${CONTAINER_NAME}$"; then
+    echo "⚠️  Container '${CONTAINER_NAME}' already exists"
+    echo ""
+    read -p "Do you want to remove it and create a new one? (y/N): " -n 1 -r
+    echo
+    if [[ $REPLY =~ ^[Yy]$ ]]; then
+        echo "🗑️  Removing existing container..."
+        docker rm -f "${CONTAINER_NAME}" 2>/dev/null || true
+    else
+        echo "Exiting..."
+        exit 0
+    fi
+fi
+
+echo "🚀 Starting Memora Standalone Container"
+echo "========================================"
+echo "Image: ${IMAGE_NAME}"
+echo "Container: ${CONTAINER_NAME}"
+echo ""
+
+# Build docker run command
+DOCKER_CMD="docker run -d --name ${CONTAINER_NAME}"
+DOCKER_CMD="${DOCKER_CMD} -p ${PORT_CONTROL}:3000"
+DOCKER_CMD="${DOCKER_CMD} -p ${PORT_API}:8080"
+DOCKER_CMD="${DOCKER_CMD} -p ${PORT_DB}:5432"
+
+if [ "$PERSIST_DATA" = true ]; then
+    DOCKER_CMD="${DOCKER_CMD} -v memora-data:/var/lib/postgresql/data"
+    echo "📦 Using persistent volume: memora-data"
+fi
+
+DOCKER_CMD="${DOCKER_CMD} ${IMAGE_NAME}"
+
+# Run the container
+eval $DOCKER_CMD
+
+echo ""
+echo "✅ Container started successfully!"
+echo ""
+echo "Services are available at:"
+echo "  - Control Plane: http://localhost:${PORT_CONTROL}"
+echo "  - Dataplane API: http://localhost:${PORT_API}"
+echo "  - PostgreSQL: localhost:${PORT_DB}"
+echo ""
+echo "View logs:"
+echo "  docker logs -f ${CONTAINER_NAME}"
+echo ""
+echo "Stop container:"
+echo "  docker stop ${CONTAINER_NAME}"
+echo ""
+echo "Remove container:"
+echo "  docker rm -f ${CONTAINER_NAME}"
+echo ""
diff --git a/standalone/supervisord.conf b/standalone/supervisord.conf
new file mode 100644
index 00000000..32b16c56
--- /dev/null
+++ b/standalone/supervisord.conf
@@ -0,0 +1,42 @@
+[supervisord]
+nodaemon=true
+user=root
+logfile=/var/log/supervisor/supervisord.log
+pidfile=/var/run/supervisord.pid
+
+[program:postgresql]
+command=/usr/lib/postgresql/15/bin/postgres -D /var/lib/postgresql/data
+user=postgres
+autostart=true
+autorestart=true
+stdout_logfile=/dev/stdout
+stdout_logfile_maxbytes=0
+stderr_logfile=/dev/stderr
+stderr_logfile_maxbytes=0
+priority=1
+
+[program:dataplane]
+command=/app/.venv/bin/python -m memora.web.server --host 0.0.0.0 --port 8080
+directory=/app/memora
+environment=PATH="/app/.venv/bin:%(ENV_PATH)s",DATABASE_URL="postgresql://postgres:postgres@localhost:5432/memora",EMBEDDING_MODEL_NAME="sentence-transformers/all-MiniLM-L6-v2",EMBEDDING_DIM="384"
+autostart=true
+autorestart=true
+stdout_logfile=/dev/stdout
+stdout_logfile_maxbytes=0
+stderr_logfile=/dev/stderr
+stderr_logfile_maxbytes=0
+startsecs=10
+priority=10
+
+[program:control-plane]
+command=/usr/bin/node /app/control-plane/server.js
+directory=/app/control-plane
+environment=NODE_ENV="production",PORT="3000",HOSTNAME="0.0.0.0",DATAPLANE_API_URL="http://localhost:8080"
+autostart=true
+autorestart=true
+stdout_logfile=/dev/stdout
+stdout_logfile_maxbytes=0
+stderr_logfile=/dev/stderr
+stderr_logfile_maxbytes=0
+startsecs=5
+priority=20
diff --git a/uv.lock b/uv.lock
index 2a01eadd..fc37f64e 100644
--- a/uv.lock
+++ b/uv.lock
@@ -10,6 +10,7 @@ resolution-markers = [
 members = [
     "benchmarks",
     "memora",
+    "memora-cli",
     "memora-dev",
 ]
 
@@ -1093,6 +1094,23 @@ requires-dist = [
     { name = "uvicorn", specifier = ">=0.38.0" },
 ]
 
+[[package]]
+name = "memora-cli"
+version = "0.1.0"
+source = { editable = "memora-cli" }
+dependencies = [
+    { name = "httpx" },
+    { name = "rich" },
+    { name = "typer" },
+]
+
+[package.metadata]
+requires-dist = [
+    { name = "httpx", specifier = ">=0.27.0" },
+    { name = "rich", specifier = ">=13.0.0" },
+    { name = "typer", specifier = ">=0.20.0" },
+]
+
 [[package]]
 name = "memora-dev"
 version = "0.1.0"