feat: Major UI/UX improvements and production readiness

## Features Added ### Document Reference System - Implemented numbered document references (@1, @2, etc.) with autocomplete dropdown - Added fuzzy filename matching for @filename references - Document filtering now prioritizes numeric refs > filename refs > all documents - Autocomplete dropdown appears when typing @ with keyboard navigation (Up/Down, Enter/Tab, Escape) - Document numbers displayed in UI for easy reference ### Conversation Management - Added conversation rename functionality with inline editing - Implemented conversation search (by title and content) - Search box always visible, even when no conversations exist - Export reports now replace @N references with actual filenames ### UI/UX Improvements - Removed debug toggle button - Improved text contrast in dark mode (better visibility) - Made input textarea expand to full available width - Fixed file text color for better readability - Enhanced document display with numbered badges ### Configuration & Timeouts - Made HTTP client timeouts configurable (connect, write, pool) - Added .env.example with all configuration options - Updated timeout documentation ### Developer Experience - Added `make test-setup` target for automated test conversation creation - Test setup script supports TEST_MESSAGE and TEST_DOCS env vars - Improved Makefile with dev and test-setup targets ### Documentation - Updated ARCHITECTURE.md with all new features - Created comprehensive deployment documentation - Added GPU VM setup guides - Removed unnecessary markdown files (CLAUDE.md, CONTRIBUTING.md, header.jpg) - Organized documentation in docs/ directory ### GPU VM / Ollama (Stability + GPU Offload) - Updated GPU VM docs to reflect the working systemd environment for remote Ollama - Standardized remote Ollama port to 11434 (and added /v1/models verification) - Documented required env for GPU offload on this VM: - `OLLAMA_MODELS=/mnt/data/ollama`, `HOME=/mnt/data/ollama/home` - `OLLAMA_LLM_LIBRARY=cuda_v12` (not `cuda`) - `LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12` ## Technical Changes ### Backend - Enhanced `docs_context.py` with reference parsing (numeric and filename) - Added `update_conversation_title` to storage.py - New endpoints: PATCH /api/conversations/{id}/title, GET /api/conversations/search - Improved report generation with filename substitution ### Frontend - Removed debugMode state and related code - Added autocomplete dropdown component - Implemented search functionality in Sidebar - Enhanced ChatInterface with autocomplete and improved textarea sizing - Updated CSS for better contrast and responsive design ## Files Changed - Backend: config.py, council.py, docs_context.py, main.py, storage.py - Frontend: App.jsx, ChatInterface.jsx, Sidebar.jsx, and related CSS files - Documentation: README.md, ARCHITECTURE.md, new docs/ directory - Configuration: .env.example, Makefile - Scripts: scripts/test_setup.py ## Breaking Changes None - all changes are backward compatible ## Testing - All existing tests pass - New test-setup script validates conversation creation workflow - Manual testing of autocomplete, search, and rename features
2025-12-28 18:15:02 -05:00
commit 3546c04348
79 changed files with 13531 additions and 0 deletions
@@ -0,0 +1,24 @@
+#!/bin/bash
+# Check if firewall might be blocking the connection
+
+GPU_VM="10.0.30.63"
+
+echo "=== Firewall Check ==="
+echo ""
+echo "The fact that Ollama shows ':::11434' means it's listening on all interfaces."
+echo "If curl still times out, it's likely a firewall issue."
+echo ""
+echo "On the GPU VM, check firewall:"
+echo "  # Check if firewall is running:"
+echo "  sudo ufw status"
+echo "  # OR"
+echo "  sudo iptables -L -n | grep 11434"
+echo ""
+echo "If firewall is blocking, allow the port:"
+echo "  sudo ufw allow 11434/tcp"
+echo "  # OR for iptables:"
+echo "  sudo iptables -A INPUT -p tcp --dport 11434 -j ACCEPT"
+echo ""
+echo "Testing connection from local machine..."
+timeout 3 curl -v http://$GPU_VM:11434/api/tags 2>&1 | grep -E "Connected|timeout|refused|Connection" | head -3
+
@@ -0,0 +1,60 @@
+#!/bin/bash
+# Check if Ollama models still exist on GPU VM
+# Run this ON THE GPU VM
+
+echo "=== Checking Ollama Models ==="
+echo ""
+
+# Check Ollama's model storage locations
+echo "1. Checking Ollama API for models:"
+curl -s http://localhost:11434/api/tags | python3 -m json.tool 2>/dev/null | grep -E '"name"|"model"' | head -10
+
+echo ""
+echo "2. Checking common model storage locations:"
+echo "   ~/.ollama/models:"
+if [ -d ~/.ollama/models ]; then
+    du -sh ~/.ollama/models
+    ls -lh ~/.ollama/models | head -5
+else
+    echo "   ✗ Not found"
+fi
+
+echo ""
+echo "   /usr/share/ollama/models:"
+if [ -d /usr/share/ollama/models ]; then
+    du -sh /usr/share/ollama/models
+    ls -lh /usr/share/ollama/models | head -5
+else
+    echo "   ✗ Not found"
+fi
+
+echo ""
+echo "   /var/lib/ollama/models:"
+if [ -d /var/lib/ollama/models ]; then
+    du -sh /var/lib/ollama/models
+    ls -lh /var/lib/ollama/models | head -5
+else
+    echo "   ✗ Not found"
+fi
+
+echo ""
+echo "3. Finding Ollama data directory:"
+if command -v ollama > /dev/null; then
+    ollama show 2>&1 | head -5
+fi
+
+echo ""
+echo "4. Checking systemd service for OLLAMA_MODELS path:"
+systemctl show ollama | grep -i model || echo "   No OLLAMA_MODELS env var set"
+
+echo ""
+echo "=== What we did ==="
+echo "We only created: /etc/systemd/system/ollama.service.d/override.conf"
+echo "This file only sets: OLLAMA_HOST=0.0.0.0:11434"
+echo "It does NOT delete models."
+echo ""
+echo "If models are missing, they might be:"
+echo "  1. In a different location (check above)"
+echo "  2. Ollama needs to be restarted to see them"
+echo "  3. Models were deleted separately (not by our script)"
+
@@ -0,0 +1,40 @@
+#!/bin/bash
+# Script to configure Ollama on GPU VM to listen on all interfaces
+# Run this on the GPU VM (10.0.30.63)
+
+echo "Configuring Ollama to listen on all interfaces..."
+
+# Method 1: Set environment variable (temporary, until reboot)
+export OLLAMA_HOST=0.0.0.0:11434
+echo "✓ Set OLLAMA_HOST=0.0.0.0:11434 (temporary)"
+
+# Method 2: Create systemd override (permanent)
+echo ""
+echo "Creating systemd override for permanent configuration..."
+sudo mkdir -p /etc/systemd/system/ollama.service.d
+sudo tee /etc/systemd/system/ollama.service.d/override.conf > /dev/null <<EOF
+[Service]
+Environment="OLLAMA_HOST=0.0.0.0:11434"
+EOF
+
+echo "✓ Created override file: /etc/systemd/system/ollama.service.d/override.conf"
+
+# Reload systemd and restart Ollama
+echo ""
+echo "Reloading systemd and restarting Ollama..."
+sudo systemctl daemon-reload
+sudo systemctl restart ollama
+
+echo ""
+echo "✓ Ollama restarted"
+echo ""
+echo "Verifying configuration..."
+sleep 2
+curl -s http://localhost:11434/api/tags | head -c 200
+echo ""
+echo ""
+echo "✓ Configuration complete!"
+echo ""
+echo "You can now test from your local machine:"
+echo "  curl http://10.0.30.63:11434/api/tags"
+
@@ -0,0 +1,57 @@
+#!/bin/bash
+# Quick diagnostic script for GPU VM connection
+
+GPU_VM="10.0.30.63"
+
+echo "=== GPU VM Connection Diagnostics ==="
+echo ""
+
+echo "1. Testing basic connectivity..."
+if ping -c 1 -W 2 $GPU_VM > /dev/null 2>&1; then
+    echo "   ✓ GPU VM is reachable"
+else
+    echo "   ✗ Cannot reach GPU VM - check network/firewall"
+    exit 1
+fi
+
+echo ""
+echo "2. Testing port 11434 (Ollama default)..."
+if timeout 3 curl -s http://$GPU_VM:11434/api/tags > /dev/null 2>&1; then
+    echo "   ✓ Port 11434 is open and responding"
+    echo "   Models available:"
+    curl -s http://$GPU_VM:11434/api/tags | python3 -m json.tool 2>/dev/null | grep -E '"name"|"model"' | head -5
+else
+    echo "   ✗ Port 11434 not responding"
+    echo "   Error details:"
+    timeout 3 curl -v http://$GPU_VM:11434/api/tags 2>&1 | grep -E "Connection|timeout|refused" | head -3
+fi
+
+echo ""
+echo "3. Testing port 8000 (alternative)..."
+if timeout 3 curl -s http://$GPU_VM:8000/v1/models > /dev/null 2>&1; then
+    echo "   ✓ Port 8000 is open and responding"
+else
+    echo "   ✗ Port 8000 not responding"
+fi
+
+echo ""
+echo "4. Checking your .env configuration..."
+if [ -f .env ]; then
+    echo "   OPENAI_COMPAT_BASE_URL: $(grep OPENAI_COMPAT_BASE_URL .env | grep -v '^#' | cut -d'=' -f2)"
+    echo "   USE_LOCAL_OLLAMA: $(grep USE_LOCAL_OLLAMA .env | grep -v '^#' | cut -d'=' -f2)"
+else
+    echo "   ✗ .env file not found"
+fi
+
+echo ""
+echo "=== Recommendations ==="
+echo ""
+echo "If port 11434 is not working:"
+echo "  1. SSH to GPU VM: ssh root@$GPU_VM"
+echo "  2. Check if Ollama is running: systemctl status ollama"
+echo "  3. Check what port Ollama is listening on: netstat -tlnp | grep ollama"
+echo "  4. If only listening on 127.0.0.1, configure it to listen on 0.0.0.0"
+echo ""
+echo "If you need to use a different port, update .env:"
+echo "  OPENAI_COMPAT_BASE_URL=http://$GPU_VM:PORT"
+
@@ -0,0 +1,59 @@
+#!/bin/bash
+# Find and verify Ollama models on GPU VM
+# Run this ON THE GPU VM
+
+echo "=== Finding Ollama Models ==="
+echo ""
+
+echo "1. Check what Ollama API reports:"
+echo "   Running: curl http://localhost:11434/api/tags"
+curl -s http://localhost:11434/api/tags | python3 -m json.tool 2>/dev/null || curl -s http://localhost:11434/api/tags
+
+echo ""
+echo ""
+echo "2. Find Ollama data directory:"
+echo "   Checking common locations..."
+
+# Check for OLLAMA_MODELS env var
+if [ -n "$OLLAMA_MODELS" ]; then
+    echo "   OLLAMA_MODELS env var: $OLLAMA_MODELS"
+    if [ -d "$OLLAMA_MODELS" ]; then
+        echo "   ✓ Found! Size: $(du -sh "$OLLAMA_MODELS" 2>/dev/null | cut -f1)"
+        echo "   Models:"
+        ls -lh "$OLLAMA_MODELS" | head -10
+    fi
+fi
+
+# Check common locations
+for dir in ~/.ollama/models ~/.ollama /usr/share/ollama/models /usr/share/ollama /var/lib/ollama/models /var/lib/ollama; do
+    if [ -d "$dir" ]; then
+        echo "   Found: $dir"
+        echo "   Size: $(du -sh "$dir" 2>/dev/null | cut -f1)"
+        if [ -d "$dir/models" ]; then
+            echo "   Models in subdirectory:"
+            ls -lh "$dir/models" 2>/dev/null | head -5
+        fi
+        find "$dir" -name "*.gguf" -o -name "*.bin" 2>/dev/null | head -5
+    fi
+done
+
+echo ""
+echo "3. Check Ollama process environment:"
+sudo cat /proc/$(pgrep -f ollama | head -1)/environ 2>/dev/null | tr '\0' '\n' | grep -i model || echo "   No OLLAMA_MODELS in process env"
+
+echo ""
+echo "4. Check systemd service environment:"
+systemctl show ollama | grep -i environment
+
+echo ""
+echo "=== If models are missing ==="
+echo "They might be in a different location. Ollama stores models in:"
+echo "  - Default: ~/.ollama/models (or /usr/share/ollama/models)"
+echo "  - Or wherever OLLAMA_MODELS env var points"
+echo ""
+echo "To re-download models:"
+echo "  ollama pull qwen2:latest"
+echo "  ollama pull qwen2.5:14b"
+echo "  ollama pull llama3.1:8b"
+echo "  ollama pull qwen2.5:7b"
+
@@ -0,0 +1,40 @@
+#!/bin/bash
+# Run this ON THE GPU VM to fix firewall
+
+echo "=== Fixing Firewall for Ollama ==="
+echo ""
+
+# Check what firewall is running
+if command -v ufw > /dev/null 2>&1; then
+    echo "Detected UFW firewall"
+    echo "Current status:"
+    sudo ufw status | head -5
+    echo ""
+    echo "Allowing port 11434..."
+    sudo ufw allow 11434/tcp
+    echo "✓ Port 11434 allowed"
+elif command -v firewall-cmd > /dev/null 2>&1; then
+    echo "Detected firewalld"
+    sudo firewall-cmd --permanent --add-port=11434/tcp
+    sudo firewall-cmd --reload
+    echo "✓ Port 11434 allowed"
+else
+    echo "Checking iptables..."
+    if sudo iptables -L -n | grep -q 11434; then
+        echo "Found iptables rules for 11434"
+    else
+        echo "Adding iptables rule..."
+        sudo iptables -A INPUT -p tcp --dport 11434 -j ACCEPT
+        echo "✓ Rule added (may need to save: sudo iptables-save)"
+    fi
+fi
+
+echo ""
+echo "Verifying Ollama is accessible..."
+sleep 1
+curl -s http://localhost:11434/api/tags | head -c 100
+echo ""
+echo ""
+echo "✓ Done! Test from your local machine:"
+echo "  curl http://10.0.30.63:11434/api/tags"
+
@@ -0,0 +1,57 @@
+#!/bin/bash
+# Run this ON THE GPU VM (10.0.30.63) to fix Ollama remote access
+
+echo "=== Fixing Ollama Remote Access ==="
+echo ""
+
+# Check if Ollama is running
+if ! systemctl is-active --quiet ollama; then
+    echo "✗ Ollama is not running. Starting it..."
+    sudo systemctl start ollama
+    sleep 2
+fi
+
+echo "✓ Ollama is running"
+echo ""
+
+# Check what it's listening on
+echo "Current Ollama listening status:"
+sudo netstat -tlnp 2>/dev/null | grep 11434 || ss -tlnp 2>/dev/null | grep 11434
+echo ""
+
+# Create systemd override
+echo "Creating systemd override to listen on all interfaces..."
+sudo mkdir -p /etc/systemd/system/ollama.service.d
+
+sudo tee /etc/systemd/system/ollama.service.d/override.conf > /dev/null <<'EOF'
+[Service]
+Environment="OLLAMA_HOST=0.0.0.0:11434"
+EOF
+
+echo "✓ Override file created"
+echo ""
+
+# Reload and restart
+echo "Reloading systemd and restarting Ollama..."
+sudo systemctl daemon-reload
+sudo systemctl restart ollama
+sleep 3
+
+echo "✓ Ollama restarted"
+echo ""
+
+# Verify
+echo "Verifying configuration..."
+if sudo netstat -tlnp 2>/dev/null | grep -q "0.0.0.0:11434" || sudo ss -tlnp 2>/dev/null | grep -q "0.0.0.0:11434"; then
+    echo "✓ SUCCESS! Ollama is now listening on 0.0.0.0:11434"
+    echo ""
+    echo "Test from your local machine:"
+    echo "  curl http://10.0.30.63:11434/api/tags"
+else
+    echo "✗ Still not listening on 0.0.0.0 - checking status..."
+    sudo systemctl status ollama --no-pager | head -10
+    echo ""
+    echo "Try checking Ollama logs:"
+    echo "  sudo journalctl -u ollama -n 20"
+fi
+
@@ -0,0 +1,87 @@
+#!/bin/bash
+# Configure Ollama to use /mnt/data for model storage
+# Run this ON THE GPU VM
+
+echo "=== Fixing Ollama Storage Location ==="
+echo ""
+
+# Check current disk usage
+echo "Current disk usage:"
+df -h | grep -E "Filesystem|/dev/sda"
+echo ""
+
+# Create models directory on /mnt/data
+echo "Creating Ollama models directory on /mnt/data..."
+sudo mkdir -p /mnt/data/ollama/models
+sudo chown -R ollama:ollama /mnt/data/ollama 2>/dev/null || sudo chown -R $(whoami):$(whoami) /mnt/data/ollama
+echo "✓ Directory created: /mnt/data/ollama/models"
+echo ""
+
+# Check if there are existing models to move
+if [ -d ~/.ollama/models ] && [ "$(ls -A ~/.ollama/models 2>/dev/null)" ]; then
+    echo "Found existing models in ~/.ollama/models"
+    echo "Moving to /mnt/data/ollama/models..."
+    sudo mv ~/.ollama/models/* /mnt/data/ollama/models/ 2>/dev/null
+    echo "✓ Models moved"
+elif [ -d /usr/share/ollama/models ] && [ "$(ls -A /usr/share/ollama/models 2>/dev/null)" ]; then
+    echo "Found existing models in /usr/share/ollama/models"
+    echo "Moving to /mnt/data/ollama/models..."
+    sudo mv /usr/share/ollama/models/* /mnt/data/ollama/models/ 2>/dev/null
+    echo "✓ Models moved"
+else
+    echo "No existing models found to move"
+fi
+
+echo ""
+
+# Update systemd service to use new location
+echo "Updating systemd service configuration..."
+sudo mkdir -p /etc/systemd/system/ollama.service.d
+
+# Check if override.conf exists and update it, or create new
+if [ -f /etc/systemd/system/ollama.service.d/override.conf ]; then
+    echo "Updating existing override.conf..."
+    # Add OLLAMA_MODELS if not already there
+    if ! grep -q "OLLAMA_MODELS" /etc/systemd/system/ollama.service.d/override.conf; then
+        sudo sed -i '/\[Service\]/a Environment="OLLAMA_MODELS=/mnt/data/ollama/models"' /etc/systemd/system/ollama.service.d/override.conf
+    else
+        sudo sed -i 's|OLLAMA_MODELS=.*|OLLAMA_MODELS=/mnt/data/ollama/models|' /etc/systemd/system/ollama.service.d/override.conf
+    fi
+else
+    echo "Creating new override.conf..."
+    sudo tee /etc/systemd/system/ollama.service.d/override.conf > /dev/null <<EOF
+[Service]
+Environment="OLLAMA_HOST=0.0.0.0:11434"
+Environment="OLLAMA_MODELS=/mnt/data/ollama/models"
+EOF
+fi
+
+echo "✓ Systemd configuration updated"
+echo ""
+
+# Reload and restart
+echo "Reloading systemd and restarting Ollama..."
+sudo systemctl daemon-reload
+sudo systemctl restart ollama
+sleep 3
+
+echo "✓ Ollama restarted with new storage location"
+echo ""
+
+# Verify
+echo "Verifying configuration:"
+echo "  Storage location: /mnt/data/ollama/models"
+echo "  Disk space available:"
+df -h /mnt/data | tail -1
+echo ""
+echo "  Checking if Ollama is running:"
+systemctl is-active ollama && echo "  ✓ Ollama is running" || echo "  ✗ Ollama is not running"
+
+echo ""
+echo "=== Done! ==="
+echo "You can now pull models:"
+echo "  ollama pull qwen2:latest"
+echo "  ollama pull qwen2.5:14b"
+echo "  ollama pull llama3.1:8b"
+echo "  ollama pull qwen2.5:7b"
+
@@ -0,0 +1,54 @@
+#!/usr/bin/env python3
+"""Quick test script to check GPU VM connection."""
+
+import asyncio
+import sys
+sys.path.insert(0, '.')
+
+from backend.llm_client import list_models
+from backend.config import OPENAI_COMPAT_BASE_URL, COUNCIL_MODELS, CHAIRMAN_MODEL
+
+async def test_connection():
+    print(f"Testing connection to: {OPENAI_COMPAT_BASE_URL}")
+    print(f"Configured council models: {COUNCIL_MODELS}")
+    print(f"Chairman model: {CHAIRMAN_MODEL}")
+    print("-" * 60)
+    
+    try:
+        models = await list_models()
+        if models is None:
+            print("✗ Unable to list models (connection error, timeout, or incompatible endpoint).")
+            print("")
+            print("Next checks:")
+            print(f"  - curl {OPENAI_COMPAT_BASE_URL.rstrip('/')}/api/tags")
+            print(f"  - curl {OPENAI_COMPAT_BASE_URL.rstrip('/')}/v1/models")
+            print("")
+            print("If you're using Ollama remotely, the port is usually 11434.")
+            return
+        if models:
+            print(f"✓ Connection successful!")
+            print(f"Found {len(models)} available models:\n")
+            for model in models:
+                marker = "✓" if model in COUNCIL_MODELS else " "
+                chairman_marker = " (CHAIRMAN)" if model == CHAIRMAN_MODEL else ""
+                print(f"  {marker} {model}{chairman_marker}")
+            
+            print("\n" + "-" * 60)
+            missing = [m for m in COUNCIL_MODELS if m not in models]
+            if missing:
+                print(f"⚠ Warning: {len(missing)} configured models not found:")
+                for m in missing:
+                    print(f"  - {m}")
+            else:
+                print("✓ All configured council models are available!")
+        else:
+            print("✗ Connected, but the server returned an empty model list.")
+            print("  This is unusual for Ollama; double-check the base URL/port and server.")
+    except Exception as e:
+        print(f"✗ Connection failed: {type(e).__name__}: {e}")
+        import traceback
+        traceback.print_exc()
+
+if __name__ == "__main__":
+    asyncio.run(test_connection())
+
@@ -0,0 +1,62 @@
+#!/usr/bin/env python3
+"""Detailed test script to check GPU VM connection and endpoints."""
+
+import asyncio
+import sys
+sys.path.insert(0, '.')
+
+import httpx
+from backend.config import OPENAI_COMPAT_BASE_URL
+
+async def test_endpoints():
+    base_url = "http://10.0.30.63"
+    ports = [8000, 11434]
+    endpoints = [
+        "/v1/models",
+        "/api/tags",
+        "/api/version",
+        "/",
+    ]
+    
+    print("Testing GPU VM connection...")
+    print("=" * 60)
+    
+    for port in ports:
+        print(f"\nTesting port {port}:")
+        print("-" * 60)
+        for endpoint in endpoints:
+            url = f"{base_url}:{port}{endpoint}"
+            try:
+                async with httpx.AsyncClient(timeout=5.0) as client:
+                    resp = await client.get(url)
+                    print(f"  {endpoint:20} -> Status: {resp.status_code}")
+                    if resp.status_code == 200:
+                        try:
+                            data = resp.json()
+                            if isinstance(data, dict):
+                                if 'data' in data:
+                                    models = data['data']
+                                    print(f"    Found {len(models)} models via /v1/models")
+                                    if models:
+                                        print(f"    First model: {models[0].get('id', models[0])}")
+                                elif 'models' in data:
+                                    models = data['models']
+                                    print(f"    Found {len(models)} models via /api/tags")
+                                    if models:
+                                        print(f"    First model: {models[0].get('name', models[0])}")
+                                else:
+                                    print(f"    Response keys: {list(data.keys())[:5]}")
+                            elif isinstance(data, list):
+                                print(f"    Found {len(data)} items")
+                        except:
+                            print(f"    Response (first 200 chars): {resp.text[:200]}")
+            except httpx.TimeoutException:
+                print(f"  {endpoint:20} -> Timeout")
+            except httpx.ConnectError:
+                print(f"  {endpoint:20} -> Connection refused")
+            except Exception as e:
+                print(f"  {endpoint:20} -> Error: {type(e).__name__}")
+
+if __name__ == "__main__":
+    asyncio.run(test_endpoints())
+
@@ -0,0 +1,48 @@
+#!/usr/bin/env python3
+"""Test that we can actually query a model and get a response."""
+
+import asyncio
+import sys
+sys.path.insert(0, '.')
+
+from backend.llm_client import query_model
+from backend.config import COUNCIL_MODELS
+
+async def test_query():
+    """Test querying one of the council models."""
+    if not COUNCIL_MODELS:
+        print("✗ No council models configured")
+        return False
+    
+    test_model = COUNCIL_MODELS[0]
+    print(f"Testing query to model: {test_model}")
+    print("-" * 60)
+    
+    try:
+        response = await query_model(
+            model=test_model,
+            messages=[{"role": "user", "content": "Say 'Hello, GPU Ollama is working!' in one sentence."}],
+            max_tokens_override=50,  # Short response for quick test
+            timeout=30.0
+        )
+        
+        if response and response.get('content'):
+            content = response['content'].strip()
+            print(f"✓ Query successful!")
+            print(f"\nResponse:")
+            print(f"  {content}")
+            return True
+        else:
+            print(f"✗ Query returned no content")
+            print(f"Response: {response}")
+            return False
+    except Exception as e:
+        print(f"✗ Query failed: {type(e).__name__}: {e}")
+        import traceback
+        traceback.print_exc()
+        return False
+
+if __name__ == "__main__":
+    success = asyncio.run(test_query())
+    sys.exit(0 if success else 1)
+
@@ -0,0 +1,70 @@
+#!/usr/bin/env python3
+"""Test script to diagnose model timeout issues."""
+import asyncio
+import time
+import httpx
+from backend.config import OPENAI_COMPAT_BASE_URL, LLM_TIMEOUT_SECONDS, DEBUG
+
+async def test_model(model: str, max_tokens: int = 10):
+    """Test a single model query."""
+    print(f"\n{'='*60}")
+    print(f"Testing model: {model}")
+    print(f"Timeout: {LLM_TIMEOUT_SECONDS}s")
+    print(f"Base URL: {OPENAI_COMPAT_BASE_URL}")
+    print(f"{'='*60}")
+    
+    url = f"{OPENAI_COMPAT_BASE_URL}/v1/chat/completions"
+    payload = {
+        "model": model,
+        "messages": [{"role": "user", "content": "Say hello"}],
+        "max_tokens": max_tokens
+    }
+    
+    start_time = time.time()
+    try:
+        async with httpx.AsyncClient(timeout=LLM_TIMEOUT_SECONDS) as client:
+            print(f"[{time.time() - start_time:.1f}s] Sending request...")
+            response = await client.post(url, json=payload)
+            elapsed = time.time() - start_time
+            print(f"[{elapsed:.1f}s] Response received: Status {response.status_code}")
+            
+            if response.status_code == 200:
+                data = response.json()
+                content = data.get("choices", [{}])[0].get("message", {}).get("content", "")
+                print(f"✓ Success! Response: {content[:100]}")
+                return True
+            else:
+                print(f"✗ Error: {response.status_code}")
+                print(f"  Response: {response.text[:200]}")
+                return False
+    except httpx.TimeoutException:
+        elapsed = time.time() - start_time
+        print(f"✗ Timeout after {elapsed:.1f}s (limit was {LLM_TIMEOUT_SECONDS}s)")
+        return False
+    except Exception as e:
+        elapsed = time.time() - start_time
+        print(f"✗ Error after {elapsed:.1f}s: {type(e).__name__}: {e}")
+        return False
+
+async def main():
+    models = ["llama3.2:1b", "qwen2.5:0.5b", "gemma2:2b"]
+    
+    print("Testing models sequentially to diagnose timeout issues...")
+    print(f"Current timeout setting: {LLM_TIMEOUT_SECONDS}s")
+    
+    results = {}
+    for model in models:
+        results[model] = await test_model(model)
+        # Small delay between tests
+        await asyncio.sleep(1)
+    
+    print(f"\n{'='*60}")
+    print("Summary:")
+    for model, success in results.items():
+        status = "✓ PASS" if success else "✗ FAIL"
+        print(f"  {model}: {status}")
+    print(f"{'='*60}")
+
+if __name__ == "__main__":
+    asyncio.run(main())
+
@@ -0,0 +1,33 @@
+#!/usr/bin/env python3
+"""Test direct connection to Ollama on GPU VM."""
+import asyncio
+import httpx
+
+async def test():
+    base = "http://10.0.30.63:11434"
+    urls = [
+        f"{base}/v1/models",
+        f"{base}/api/tags",
+    ]
+    
+    for url in urls:
+        print(f"\nTesting: {url}")
+        try:
+            async with httpx.AsyncClient(timeout=5.0) as client:
+                resp = await client.get(url)
+                print(f"  Status: {resp.status_code}")
+                if resp.status_code == 200:
+                    data = resp.json()
+                    print(f"  Keys: {list(data.keys())}")
+                    if 'models' in data:
+                        models = [m.get('name', m.get('model', m)) for m in data['models']]
+                        print(f"  Found {len(models)} models: {models}")
+                    elif 'data' in data:
+                        models = [m.get('id', m) for m in data['data']]
+                        print(f"  Found {len(models)} models: {models}")
+        except httpx.TimeoutException:
+            print(f"  ✗ Timeout - Ollama may only be listening on localhost")
+        except Exception as e:
+            print(f"  ✗ Error: {e}")
+
+asyncio.run(test())
@@ -0,0 +1,140 @@
+#!/usr/bin/env python3
+"""Setup a test conversation with optional message and documents."""
+
+import os
+import sys
+import httpx
+import time
+from datetime import datetime
+from pathlib import Path
+from dotenv import load_dotenv
+from urllib.parse import quote
+
+# Load .env file if it exists
+load_dotenv()
+
+API_BASE = "http://localhost:8001"
+
+def create_test_conversation():
+    """Create a new conversation with today's date/time as title."""
+    now = datetime.now()
+    title = now.strftime("%Y-%m-%d %H:%M:%S")
+    
+    with httpx.Client(timeout=10.0) as client:
+        # Create conversation
+        response = client.post(f"{API_BASE}/api/conversations", json={})
+        if response.status_code != 200:
+            print(f"Error creating conversation: {response.text}", file=sys.stderr)
+            sys.exit(1)
+        
+        conv = response.json()
+        conv_id = conv["id"]
+        
+        # Update title
+        response = client.patch(
+            f"{API_BASE}/api/conversations/{conv_id}/title",
+            json={"title": title}
+        )
+        if response.status_code != 200:
+            print(f"Warning: Could not set title: {response.text}", file=sys.stderr)
+        
+        print(f"Created conversation: {conv_id}")
+        print(f"Title: {title}")
+        return conv_id
+
+def upload_document(conv_id, filepath):
+    """Upload a document to a conversation."""
+    path = Path(filepath)
+    if not path.exists():
+        print(f"Warning: File not found: {filepath}", file=sys.stderr)
+        return False
+    
+    with httpx.Client(timeout=30.0) as client:
+        with open(path, 'rb') as f:
+            files = {'file': (path.name, f.read(), 'text/markdown')}
+            response = client.post(
+                f"{API_BASE}/api/conversations/{conv_id}/documents",
+                files=files
+            )
+            if response.status_code != 200:
+                print(f"Warning: Could not upload {filepath}: {response.text}", file=sys.stderr)
+                return False
+    
+    print(f"Uploaded: {path.name}")
+    return True
+
+def send_message(conv_id, message):
+    """Send a message to a conversation."""
+    # Use a very long timeout since the council process can take several minutes
+    # Default to 10 minutes, but allow override via env var
+    timeout_seconds = float(os.getenv("TEST_MESSAGE_TIMEOUT_SECONDS", "600.0"))
+    
+    print(f"Sending message (this may take several minutes, timeout: {timeout_seconds}s)...")
+    with httpx.Client(timeout=timeout_seconds) as client:
+        try:
+            response = client.post(
+                f"{API_BASE}/api/conversations/{conv_id}/message",
+                json={"content": message}
+            )
+            if response.status_code != 200:
+                print(f"Warning: Could not send message: {response.text}", file=sys.stderr)
+                return False
+            print(f"✓ Message sent and processed: {message[:50]}...")
+            return True
+        except httpx.ReadTimeout:
+            print(f"Error: Request timed out after {timeout_seconds}s", file=sys.stderr)
+            print("The council process is still running. You can check the conversation in the UI.", file=sys.stderr)
+            print(f"Conversation ID: {conv_id}", file=sys.stderr)
+            return False
+        except httpx.RequestError as e:
+            print(f"Error sending message: {e}", file=sys.stderr)
+            return False
+
+def main():
+    # Wait for backend to be ready
+    max_retries = 10
+    for i in range(max_retries):
+        try:
+            with httpx.Client(timeout=1.0) as client:
+                response = client.get(f"{API_BASE}/")
+                if response.status_code == 200:
+                    break
+        except httpx.RequestError:
+            if i < max_retries - 1:
+                time.sleep(1)
+            else:
+                print(f"Error: Backend not available at {API_BASE}", file=sys.stderr)
+                print("Make sure the backend is running: uv run python -m backend.main", file=sys.stderr)
+                sys.exit(1)
+    
+    # Create conversation
+    conv_id = create_test_conversation()
+    
+    # Upload documents if specified
+    test_docs = os.getenv("TEST_DOCS", "")
+    if test_docs:
+        doc_paths = [p.strip() for p in test_docs.split(",") if p.strip()]
+        for doc_path in doc_paths:
+            upload_document(conv_id, doc_path)
+    
+    # Note: TEST_MESSAGE is NOT sent automatically
+    # It's provided here for reference - user should type it in the UI
+    test_message = os.getenv("TEST_MESSAGE", "")
+    open_url = f"http://localhost:5173/?conversation={conv_id}"
+    if test_message:
+        open_url += f"&message={quote(test_message)}"
+    
+    print(f"\n✓ Conversation created: {conv_id}")
+    print(f"CONVERSATION_ID={conv_id}")  # For Makefile to parse
+    print(f"OPEN_URL={open_url}")  # For Makefile to parse
+    print(f"Open in browser: {open_url}")
+    
+    if test_message:
+        print(f"\n💡 Pre-filled message (copy/paste into input):")
+        print(f"   {test_message}")
+    
+    return conv_id
+
+if __name__ == "__main__":
+    main()
+