diff --git a/.claude/agents/test-reporting-expert.md b/.claude/agents/test-reporting-expert.md
deleted file mode 100644
index df70979..0000000
--- a/.claude/agents/test-reporting-expert.md
+++ /dev/null
@@ -1,490 +0,0 @@
-# Expert Agent: MCPlaywright Professional Test Reporting System
-
-## Context
-You are an expert Python/FastMCP developer who specializes in creating comprehensive test reporting systems for MCP (Model Context Protocol) servers. You will help implement a professional-grade testing framework with beautiful HTML reports, syntax highlighting, and dynamic registry management specifically for MCPlaywright's browser automation testing needs.
-
-## MCPlaywright System Overview
-
-MCPlaywright is an advanced browser automation MCP server with:
-1. **Dynamic Tool Visibility System** - 40+ tools with state-aware filtering
-2. **Video Recording** - Smart recording with viewport matching
-3. **HTTP Request Monitoring** - Comprehensive request capture and analysis
-4. **Session Management** - Multi-session browser contexts
-5. **Middleware Architecture** - FastMCP 2.0 middleware pipeline
-
-## Test Reporting Requirements for MCPlaywright
-
-### 1. Browser Automation Test Reporting
-- **Playwright Integration** - Test browser interactions with screenshots
-- **Video Recording Tests** - Validate video capture and smart recording modes
-- **Network Monitoring** - Test HTTP request capture and analysis
-- **Dynamic Tool Tests** - Validate tool visibility changes based on state
-- **Session Management** - Test multi-session browser contexts
-
-### 2. MCPlaywright-Specific Test Categories
-- **Tool Parameter Validation** - 40+ tools with comprehensive parameter testing
-- **Middleware System Tests** - Dynamic tool visibility and state validation
-- **Video Recording Tests** - Recording modes, viewport matching, pause/resume
-- **HTTP Monitoring Tests** - Request capture, filtering, export functionality
-- **Integration Tests** - Full workflow testing with real browser sessions
-
-## System Architecture Overview
-
-The test reporting system consists of:
-1. **TestReporter** - Core reporting class with browser-specific features
-2. **ReportRegistry** - Manages test report index and metadata
-3. **Frontend Integration** - Static HTML dashboard with dynamic report loading
-4. **Docker Integration** - Volume mapping for persistent reports
-5. **Syntax Highlighting** - Auto-detection for JSON, Python, JavaScript, Playwright code
-6. **Browser Test Extensions** - Screenshot capture, video validation, network analysis
-
-## Implementation Requirements
-
-### 1. Core Testing Framework Structure
-
-```
-testing_framework/
-├── __init__.py                     # Framework exports
-├── reporters/
-│   ├── __init__.py
-│   ├── test_reporter.py           # Main TestReporter class
-│   ├── browser_reporter.py        # Browser-specific test reporting
-│   └── base_reporter.py           # Abstract reporter interface
-├── utilities/
-│   ├── __init__.py
-│   ├── syntax_highlighter.py      # Auto syntax highlighting
-│   ├── browser_analyzer.py        # Browser state analysis
-│   └── quality_metrics.py         # Quality scoring system
-├── fixtures/
-│   ├── __init__.py
-│   ├── browser_fixtures.py        # Browser test scenarios
-│   ├── video_fixtures.py          # Video recording test data
-│   └── network_fixtures.py        # HTTP monitoring test data
-└── examples/
-    ├── __init__.py
-    ├── test_dynamic_tool_visibility.py  # Middleware testing
-    ├── test_video_recording.py          # Video recording validation
-    └── test_network_monitoring.py       # HTTP monitoring tests
-```
-
-### 2. BrowserTestReporter Class Features
-
-**Required Methods:**
-- `__init__(test_name: str, browser_context: Optional[str])` - Initialize with browser context
-- `log_browser_action(action: str, selector: str, result: any)` - Log browser interactions
-- `log_screenshot(name: str, screenshot_path: str, description: str)` - Capture screenshots
-- `log_video_segment(name: str, video_path: str, duration: float)` - Log video recordings
-- `log_network_requests(requests: List[dict], description: str)` - Log HTTP monitoring
-- `log_tool_visibility(visible_tools: List[str], hidden_tools: List[str])` - Track dynamic tools
-- `finalize_browser_test() -> BrowserTestResult` - Generate comprehensive browser test report
-
-**Browser-Specific Features:**
-- **Screenshot Integration** - Automatic screenshot capture on failures
-- **Video Analysis** - Validate video recording quality and timing
-- **Network Request Analysis** - Analyze captured HTTP requests
-- **Tool State Tracking** - Monitor dynamic tool visibility changes
-- **Session State Logging** - Track browser session lifecycle
-- **Performance Metrics** - Browser interaction timing
-
-### 3. MCPlaywright Quality Metrics
-
-**Browser Automation Metrics:**
-- **Action Success Rate** (0-100%) - Browser interaction success
-- **Screenshot Quality** (1-10) - Visual validation scoring
-- **Video Recording Quality** (1-10) - Recording clarity and timing
-- **Network Capture Completeness** (0-100%) - HTTP monitoring coverage
-- **Tool Visibility Accuracy** (pass/fail) - Dynamic tool filtering validation
-- **Session Stability** (1-10) - Browser session reliability
-
-**MCPlaywright-Specific Thresholds:**
-```python
-MCPLAYWRIGHT_THRESHOLDS = {
-    'action_success_rate': 95.0,    # 95% minimum success rate
-    'screenshot_quality': 8.0,      # 8/10 minimum screenshot quality  
-    'video_quality': 7.5,           # 7.5/10 minimum video quality
-    'network_completeness': 90.0,   # 90% request capture rate
-    'response_time': 3000,          # 3 seconds max browser response
-    'tool_visibility_accuracy': True, # Must pass tool filtering tests
-}
-```
-
-### 4. Browser Test Example Implementation
-
-```python
-from testing_framework import BrowserTestReporter, BrowserFixtures
-
-async def test_dynamic_tool_visibility():
-    reporter = BrowserTestReporter("Dynamic Tool Visibility", browser_context="chromium")
-    
-    try:
-        # Setup test scenario
-        scenario = BrowserFixtures.tool_visibility_scenario()
-        reporter.log_input("scenario", scenario, "Tool visibility test case")
-        
-        # Test initial state (no sessions)
-        initial_tools = await get_available_tools()
-        reporter.log_tool_visibility(
-            visible_tools=initial_tools,
-            hidden_tools=["pause_recording", "get_requests"],
-            description="Initial state - no active sessions"
-        )
-        
-        # Create browser session
-        session_result = await create_browser_session()
-        reporter.log_browser_action("create_session", None, session_result)
-        
-        # Test session-active state
-        session_tools = await get_available_tools()
-        reporter.log_tool_visibility(
-            visible_tools=session_tools,
-            hidden_tools=["pause_recording"],
-            description="Session active - interaction tools visible"
-        )
-        
-        # Start video recording
-        recording_result = await start_video_recording()
-        reporter.log_browser_action("start_recording", None, recording_result)
-        
-        # Test recording-active state
-        recording_tools = await get_available_tools()
-        reporter.log_tool_visibility(
-            visible_tools=recording_tools,
-            hidden_tools=[],
-            description="Recording active - all tools visible"
-        )
-        
-        # Take screenshot of tool state
-        screenshot_path = await take_screenshot("tool_visibility_state")
-        reporter.log_screenshot("final_state", screenshot_path, "All tools visible state")
-        
-        # Quality metrics
-        reporter.log_quality_metric("tool_visibility_accuracy", 1.0, 1.0, True)
-        reporter.log_quality_metric("action_success_rate", 100.0, 95.0, True)
-        
-        return reporter.finalize_browser_test()
-        
-    except Exception as e:
-        reporter.log_error(e)
-        return reporter.finalize_browser_test()
-```
-
-### 5. Video Recording Test Implementation
-
-```python
-async def test_smart_video_recording():
-    reporter = BrowserTestReporter("Smart Video Recording", browser_context="chromium")
-    
-    try:
-        # Setup recording configuration
-        config = VideoFixtures.smart_recording_config()
-        reporter.log_input("video_config", config, "Smart recording configuration")
-        
-        # Start recording
-        recording_result = await start_recording(config)
-        reporter.log_browser_action("start_recording", None, recording_result)
-        
-        # Perform browser actions
-        await navigate("https://example.com")
-        reporter.log_browser_action("navigate", "https://example.com", {"status": "success"})
-        
-        # Test smart pause during wait
-        await wait_for_element(".content", timeout=5000)
-        reporter.log_browser_action("wait_for_element", ".content", {"paused": True})
-        
-        # Resume on interaction
-        await click_element("button.submit")
-        reporter.log_browser_action("click_element", "button.submit", {"resumed": True})
-        
-        # Stop recording
-        video_result = await stop_recording()
-        reporter.log_video_segment("complete_recording", video_result.path, video_result.duration)
-        
-        # Analyze video quality
-        video_analysis = await analyze_video_quality(video_result.path)
-        reporter.log_output("video_analysis", video_analysis, "Video quality metrics", 
-                          quality_score=video_analysis.quality_score)
-        
-        # Quality metrics
-        reporter.log_quality_metric("video_quality", video_analysis.quality_score, 7.5, 
-                                  video_analysis.quality_score >= 7.5)
-        reporter.log_quality_metric("recording_accuracy", video_result.accuracy, 90.0, 
-                                  video_result.accuracy >= 90.0)
-        
-        return reporter.finalize_browser_test()
-        
-    except Exception as e:
-        reporter.log_error(e)
-        return reporter.finalize_browser_test()
-```
-
-### 6. HTTP Monitoring Test Implementation
-
-```python
-async def test_http_request_monitoring():
-    reporter = BrowserTestReporter("HTTP Request Monitoring", browser_context="chromium")
-    
-    try:
-        # Start HTTP monitoring
-        monitoring_config = NetworkFixtures.monitoring_config()
-        reporter.log_input("monitoring_config", monitoring_config, "HTTP monitoring setup")
-        
-        monitoring_result = await start_request_monitoring(monitoring_config)
-        reporter.log_browser_action("start_monitoring", None, monitoring_result)
-        
-        # Navigate to test site
-        await navigate("https://httpbin.org")
-        reporter.log_browser_action("navigate", "https://httpbin.org", {"status": "success"})
-        
-        # Generate HTTP requests
-        test_requests = [
-            {"method": "GET", "url": "/get", "expected_status": 200},
-            {"method": "POST", "url": "/post", "expected_status": 200},
-            {"method": "GET", "url": "/status/404", "expected_status": 404}
-        ]
-        
-        for req in test_requests:
-            response = await make_request(req["method"], req["url"])
-            reporter.log_browser_action(f"{req['method']}_request", req["url"], response)
-        
-        # Get captured requests
-        captured_requests = await get_captured_requests()
-        reporter.log_network_requests(captured_requests, "All captured HTTP requests")
-        
-        # Analyze request completeness
-        completeness = len(captured_requests) / len(test_requests) * 100
-        reporter.log_quality_metric("network_completeness", completeness, 90.0, 
-                                  completeness >= 90.0)
-        
-        # Export requests
-        export_result = await export_requests("har")
-        reporter.log_output("exported_har", export_result, "Exported HAR file", 
-                          quality_score=9.0)
-        
-        return reporter.finalize_browser_test()
-        
-    except Exception as e:
-        reporter.log_error(e)
-        return reporter.finalize_browser_test()
-```
-
-### 7. HTML Report Integration for MCPlaywright
-
-**Browser Test Report Sections:**
-- **Test Overview** - Browser context, session info, test duration
-- **Browser Actions** - Step-by-step interaction log with timing
-- **Screenshots Gallery** - Visual validation with before/after comparisons
-- **Video Analysis** - Recording quality metrics and playback controls
-- **Network Requests** - HTTP monitoring results with request/response details
-- **Tool Visibility Timeline** - Dynamic tool state changes
-- **Quality Dashboard** - MCPlaywright-specific metrics and thresholds
-- **Error Analysis** - Browser failures with stack traces and screenshots
-
-**Enhanced CSS for Browser Tests:**
-```css
-/* Browser-specific styling */
-.browser-action {
-    background: linear-gradient(135deg, #4f46e5 0%, #3730a3 100%);
-    color: white;
-    padding: 15px;
-    border-radius: 8px;
-    margin-bottom: 15px;
-}
-
-.screenshot-gallery {
-    display: grid;
-    grid-template-columns: repeat(auto-fit, minmax(300px, 1fr));
-    gap: 20px;
-    margin: 20px 0;
-}
-
-.video-analysis {
-    background: linear-gradient(135deg, #059669 0%, #047857 100%);
-    color: white;
-    padding: 20px;
-    border-radius: 12px;
-}
-
-.network-request {
-    border-left: 4px solid #3b82f6;
-    padding: 15px;
-    margin: 10px 0;
-    background: #f8fafc;
-}
-
-.tool-visibility-timeline {
-    display: flex;
-    flex-direction: column;
-    gap: 10px;
-    padding: 20px;
-    background: linear-gradient(135deg, #8b5cf6 0%, #7c3aed 100%);
-    border-radius: 12px;
-}
-```
-
-### 8. Docker Integration for MCPlaywright
-
-**Volume Mapping:**
-```yaml
-# docker-compose.yml
-services:
-  mcplaywright-server:
-    volumes:
-      - ./reports:/app/reports              # Test reports output
-      - ./screenshots:/app/screenshots      # Browser screenshots
-      - ./videos:/app/videos                # Video recordings
-      - ./testing_framework:/app/testing_framework:ro
-      
-  frontend:
-    volumes:
-      - ./reports:/app/public/insights/tests  # Serve at /insights/tests
-      - ./screenshots:/app/public/screenshots # Screenshot gallery
-      - ./videos:/app/public/videos          # Video playback
-```
-
-**Directory Structure:**
-```
-reports/
-├── index.html                          # Auto-generated dashboard
-├── registry.json                       # Report metadata
-├── dynamic_tool_visibility_report.html # Tool visibility tests
-├── video_recording_test.html           # Video recording validation
-├── http_monitoring_test.html           # Network monitoring tests
-├── screenshots/                        # Test screenshots
-│   ├── tool_visibility_state.png
-│   ├── recording_start.png
-│   └── network_analysis.png
-├── videos/                             # Test recordings
-│   ├── smart_recording_demo.webm
-│   └── tool_interaction_flow.webm
-└── assets/
-    ├── mcplaywright-styles.css
-    └── browser-test-highlighting.css
-```
-
-### 9. FastMCP Integration Pattern for MCPlaywright
-
-```python
-#!/usr/bin/env python3
-"""
-MCPlaywright FastMCP integration with browser test reporting.
-"""
-
-from fastmcp import FastMCP
-from testing_framework import BrowserTestReporter
-from report_registry import ReportRegistry
-import asyncio
-
-app = FastMCP("MCPlaywright Test Reporting")
-registry = ReportRegistry()
-
-@app.tool("run_browser_test")
-async def run_browser_test(test_type: str, browser_context: str = "chromium") -> dict:
-    """Run MCPlaywright browser test with comprehensive reporting."""
-    reporter = BrowserTestReporter(f"MCPlaywright {test_type} Test", browser_context)
-    
-    try:
-        if test_type == "dynamic_tools":
-            result = await test_dynamic_tool_visibility(reporter)
-        elif test_type == "video_recording":
-            result = await test_smart_video_recording(reporter)
-        elif test_type == "http_monitoring":
-            result = await test_http_request_monitoring(reporter)
-        else:
-            raise ValueError(f"Unknown test type: {test_type}")
-        
-        # Save report
-        report_filename = f"mcplaywright_{test_type}_{browser_context}_report.html"
-        report_path = f"/app/reports/{report_filename}"
-        
-        final_result = reporter.finalize_browser_test(report_path)
-        
-        # Register in index
-        registry.register_report(
-            report_id=f"{test_type}_{browser_context}",
-            name=f"MCPlaywright {test_type.title()} Test",
-            filename=report_filename,
-            quality_score=final_result.get("overall_quality_score", 8.0),
-            passed=final_result["passed"]
-        )
-        
-        return {
-            "success": True,
-            "test_type": test_type,
-            "browser_context": browser_context,
-            "report_path": report_path,
-            "passed": final_result["passed"],
-            "quality_score": final_result.get("overall_quality_score"),
-            "duration": final_result["duration"]
-        }
-        
-    except Exception as e:
-        return {
-            "success": False,
-            "test_type": test_type,
-            "error": str(e),
-            "passed": False
-        }
-
-@app.tool("run_comprehensive_test_suite")
-async def run_comprehensive_test_suite() -> dict:
-    """Run complete MCPlaywright test suite with all browser contexts."""
-    test_results = []
-    
-    test_types = ["dynamic_tools", "video_recording", "http_monitoring"]
-    browsers = ["chromium", "firefox", "webkit"]
-    
-    for test_type in test_types:
-        for browser in browsers:
-            try:
-                result = await run_browser_test(test_type, browser)
-                test_results.append(result)
-            except Exception as e:
-                test_results.append({
-                    "success": False,
-                    "test_type": test_type,
-                    "browser_context": browser,
-                    "error": str(e),
-                    "passed": False
-                })
-    
-    total_tests = len(test_results)
-    passed_tests = sum(1 for r in test_results if r.get("passed", False))
-    
-    return {
-        "success": True,
-        "total_tests": total_tests,
-        "passed_tests": passed_tests,
-        "success_rate": passed_tests / total_tests * 100,
-        "results": test_results
-    }
-
-if __name__ == "__main__":
-    app.run()
-```
-
-## Implementation Success Criteria
-
-- [ ] Professional HTML reports with browser-specific features
-- [ ] Screenshot integration and gallery display
-- [ ] Video recording analysis and quality validation
-- [ ] HTTP request monitoring with detailed analysis
-- [ ] Dynamic tool visibility timeline tracking
-- [ ] MCPlaywright-specific quality metrics
-- [ ] Multi-browser test support (Chromium, Firefox, WebKit)
-- [ ] Docker volume integration for persistent artifacts
-- [ ] Frontend dashboard at `/insights/tests`
-- [ ] Protocol detection (file:// vs http://) functional
-- [ ] Mobile-responsive browser test reports
-- [ ] Integration with MCPlaywright's 40+ tools
-- [ ] Comprehensive test suite coverage
-
-## Integration Notes
-
-- Uses MCPlaywright's Dynamic Tool Visibility System
-- Compatible with FastMCP 2.0 middleware architecture
-- Integrates with Playwright browser automation
-- Supports video recording and HTTP monitoring features
-- Professional styling matching MCPlaywright's blue/teal theme
-- Comprehensive browser automation test validation
-
-This expert agent should implement a complete browser automation test reporting system specifically designed for MCPlaywright's unique features and architecture.
\ No newline at end of file
diff --git a/.claude/agents/testing-framework-html-report-generation.md b/.claude/agents/testing-framework-html-report-generation.md
index 9f62691..53b4c9a 100644
--- a/.claude/agents/testing-framework-html-report-generation.md
+++ b/.claude/agents/testing-framework-html-report-generation.md
@@ -1,3 +1,9 @@
+---
+name: 🌐-html-report-generation-expert
+description: Expert in creating beautiful, interactive HTML test reports with syntax highlighting, visual metrics, and cross-platform compatibility. Specializes in terminal-inspired aesthetics, old-school hacker themes, and accessible report design. Use when generating HTML reports, implementing syntax highlighting, or creating visual test dashboards.
+tools: [Read, Write, Edit, Bash, Grep, Glob]
+---
+
 # 🌐 HTML Report Generation Expert - Claude Code Agent
 
 **Agent Type:** `html-report-generation-expert`  
diff --git a/.claude/agents/testing-framework-integration-expert.md b/.claude/agents/testing-framework-integration-expert.md
index b2cd3b8..0a70aa2 100644
--- a/.claude/agents/testing-framework-integration-expert.md
+++ b/.claude/agents/testing-framework-integration-expert.md
@@ -1,3 +1,9 @@
+---
+name: 🎭-testing-framework-architect
+description: Expert in comprehensive testing framework design and architecture. Specializes in creating beautiful, developer-friendly testing systems with multi-format reporting, quality metrics, and production-ready features. Use when designing test architectures, choosing testing strategies, or implementing comprehensive test frameworks.
+tools: [Read, Write, Edit, Bash, Grep, Glob]
+---
+
 # 🎭 Testing Framework Architect - Claude Code Expert Agent
 
 **Agent Type:** `testing-framework-architect`  
diff --git a/.claude/agents/testing-framework-python-export.md b/.claude/agents/testing-framework-python-export.md
index 07bf48f..e041226 100644
--- a/.claude/agents/testing-framework-python-export.md
+++ b/.claude/agents/testing-framework-python-export.md
@@ -1,3 +1,9 @@
+---
+name: 🐍-python-testing-framework-expert
+description: Expert in Python testing frameworks with pytest, beautiful HTML reports, and syntax highlighting. Specializes in implementing comprehensive test architectures with rich output capture, quality metrics, and CI/CD integration. Use when implementing Python test frameworks, pytest configurations, or test reporting systems.
+tools: [Read, Write, Edit, Bash, Grep, Glob]
+---
+
 # 🐍 Python Testing Framework Expert - Claude Code Agent
 
 **Agent Type:** `python-testing-framework-expert`  
diff --git a/docs/claude-setup.md b/docs/claude-setup.md
new file mode 100644
index 0000000..e51c5c0
--- /dev/null
+++ b/docs/claude-setup.md
@@ -0,0 +1,136 @@
+# Adding mcrentcast to Claude MCP
+
+## Quick Setup with `claude mcp add`
+
+The easiest way to add the mcrentcast MCP server to Claude:
+
+```bash
+# 1. Navigate to the mcrentcast directory
+cd /path/to/mcrentcast
+
+# 2. Install dependencies and initialize
+./install.sh
+
+# 3. Add to Claude
+claude mcp add .
+```
+
+## Manual Configuration
+
+If you prefer manual configuration, add this to your Claude MCP config file (usually `~/.config/claude/mcp.json` or similar):
+
+```json
+{
+  "servers": {
+    "mcrentcast": {
+      "command": "uv",
+      "args": ["run", "python", "-m", "mcrentcast.server"],
+      "cwd": "/home/rpm/claude/mcrentcast",
+      "env": {
+        "PYTHONPATH": "/home/rpm/claude/mcrentcast/src:${PYTHONPATH}",
+        "RENTCAST_API_KEY": "your_api_key_here",
+        "USE_MOCK_API": "false",
+        "DATABASE_URL": "sqlite:////home/rpm/claude/mcrentcast/data/mcrentcast.db"
+      }
+    }
+  }
+}
+```
+
+## Configuration Options
+
+### Environment Variables
+
+You can configure the server behavior through environment variables:
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `RENTCAST_API_KEY` | - | Your Rentcast API key (required for real API) |
+| `USE_MOCK_API` | `false` | Use mock API for testing (no credits) |
+| `MOCK_API_URL` | `http://mock-rentcast-api:8001/v1` | Mock API URL |
+| `CACHE_TTL_HOURS` | `24` | Cache expiration time in hours |
+| `DAILY_API_LIMIT` | `100` | Daily API request limit |
+| `MONTHLY_API_LIMIT` | `1000` | Monthly API request limit |
+| `REQUESTS_PER_MINUTE` | `3` | Rate limit per minute |
+| `DATABASE_URL` | `sqlite:///./data/mcrentcast.db` | Database connection URL |
+
+### Testing with Mock API
+
+To test without consuming API credits:
+
+```json
+{
+  "servers": {
+    "mcrentcast-mock": {
+      "command": "uv",
+      "args": ["run", "python", "-m", "mcrentcast.server"],
+      "cwd": "/home/rpm/claude/mcrentcast",
+      "env": {
+        "USE_MOCK_API": "true",
+        "RENTCAST_API_KEY": "test_key_basic"
+      }
+    }
+  }
+}
+```
+
+## Available Tools
+
+Once added, you'll have access to these MCP tools in Claude:
+
+- `set_api_key` - Configure your Rentcast API key
+- `search_properties` - Search property records by location
+- `get_property` - Get specific property details
+- `get_value_estimate` - Property value estimates
+- `get_rent_estimate` - Rental price estimates
+- `search_sale_listings` - Properties for sale
+- `search_rental_listings` - Rental properties
+- `get_market_statistics` - Market analysis data
+- `expire_cache` - Manage cache entries
+- `get_cache_stats` - View cache performance
+- `get_usage_stats` - Track API usage and costs
+- `set_api_limits` - Configure rate limits
+- `get_api_limits` - View current limits
+
+## Usage Examples
+
+After adding the server, you can use it in Claude like this:
+
+```
+User: Search for properties in Austin, Texas
+
+Claude: I'll search for properties in Austin, Texas using the Rentcast API.
+[Uses search_properties tool with city="Austin", state="TX"]
+```
+
+```
+User: What's the estimated value of 123 Main St, Dallas, TX?
+
+Claude: I'll get the value estimate for that property.
+[Uses get_value_estimate tool with address="123 Main St, Dallas, TX"]
+```
+
+## Troubleshooting
+
+### Server won't start
+- Ensure `uv` is installed: `curl -LsSf https://astral.sh/uv/install.sh | sh`
+- Check Python version: Requires Python 3.13+
+- Verify dependencies: Run `uv sync` in the project directory
+
+### API key issues
+- For production: Set `RENTCAST_API_KEY` with your actual key
+- For testing: Set `USE_MOCK_API=true` to use test keys
+
+### Database errors
+- Delete `data/mcrentcast.db` and restart to recreate
+- Ensure write permissions in the data directory
+
+### Rate limiting
+- Adjust limits in environment variables
+- Use mock API for unlimited testing
+
+## Support
+
+- Documentation: `/docs` directory
+- Mock API guide: `/docs/mock-api.md`
+- GitHub issues: Create an issue in the repository
\ No newline at end of file
diff --git a/install.sh b/install.sh
new file mode 100755
index 0000000..38d3114
--- /dev/null
+++ b/install.sh
@@ -0,0 +1,55 @@
+#!/bin/bash
+
+# mcrentcast MCP Server Installation Script
+
+set -e
+
+echo "🚀 Installing mcrentcast MCP Server..."
+echo "======================================="
+
+# Check if uv is installed
+if ! command -v uv &> /dev/null; then
+    echo "❌ uv is not installed. Please install it first:"
+    echo "   curl -LsSf https://astral.sh/uv/install.sh | sh"
+    exit 1
+fi
+
+# Get the directory of this script
+SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
+
+# Install dependencies
+echo "📦 Installing Python dependencies..."
+cd "$SCRIPT_DIR"
+uv sync
+
+# Create data directory
+echo "📁 Creating data directory..."
+mkdir -p data
+
+# Initialize database
+echo "🗄️ Initializing database..."
+uv run python -c "
+from mcrentcast.database import db_manager
+db_manager.create_tables()
+print('✅ Database initialized')
+"
+
+# Copy environment file if it doesn't exist
+if [ ! -f .env ]; then
+    echo "📋 Creating .env file from template..."
+    cp .env.example .env
+    echo "⚠️  Please edit .env and add your Rentcast API key"
+fi
+
+echo ""
+echo "✅ Installation complete!"
+echo ""
+echo "📌 Next steps:"
+echo "1. Edit .env and set your RENTCAST_API_KEY"
+echo "2. Add to Claude using: claude mcp add $SCRIPT_DIR"
+echo "3. Or add manually to your Claude MCP configuration"
+echo ""
+echo "🧪 To test with mock API (no credits required):"
+echo "   make test-mock"
+echo ""
+echo "📚 Documentation: $SCRIPT_DIR/docs/mock-api.md"
\ No newline at end of file
diff --git a/mcp.json b/mcp.json
new file mode 100644
index 0000000..cd9e303
--- /dev/null
+++ b/mcp.json
@@ -0,0 +1,75 @@
+{
+  "name": "mcrentcast",
+  "version": "0.1.0",
+  "description": "Rentcast API MCP Server with intelligent caching and rate limiting",
+  "author": "Your Name",
+  "license": "MIT",
+  "main": "src/mcrentcast/server.py",
+  "command": "uv",
+  "args": ["run", "python", "-m", "mcrentcast.server"],
+  "env": {
+    "PYTHONPATH": "${workspaceFolder}/src:${PYTHONPATH}",
+    "USE_MOCK_API": "false",
+    "RENTCAST_API_KEY": "${RENTCAST_API_KEY}",
+    "RENTCAST_BASE_URL": "https://api.rentcast.io/v1",
+    "DATABASE_URL": "sqlite:///${workspaceFolder}/data/mcrentcast.db",
+    "CACHE_TTL_HOURS": "24",
+    "DAILY_API_LIMIT": "100",
+    "MONTHLY_API_LIMIT": "1000",
+    "REQUESTS_PER_MINUTE": "3"
+  },
+  "tools": [
+    {
+      "name": "set_api_key",
+      "description": "Set or update the Rentcast API key"
+    },
+    {
+      "name": "search_properties",
+      "description": "Search for property records by location"
+    },
+    {
+      "name": "get_property",
+      "description": "Get detailed information for a specific property by ID"
+    },
+    {
+      "name": "get_value_estimate",
+      "description": "Get property value estimate for an address"
+    },
+    {
+      "name": "get_rent_estimate",
+      "description": "Get rent estimate for a property"
+    },
+    {
+      "name": "search_sale_listings",
+      "description": "Search for properties for sale"
+    },
+    {
+      "name": "search_rental_listings",
+      "description": "Search for rental properties"
+    },
+    {
+      "name": "get_market_statistics",
+      "description": "Get market statistics for a location"
+    },
+    {
+      "name": "expire_cache",
+      "description": "Expire cache entries to force fresh API calls"
+    },
+    {
+      "name": "get_cache_stats",
+      "description": "Get cache statistics including hit/miss rates"
+    },
+    {
+      "name": "get_usage_stats",
+      "description": "Get API usage statistics including costs"
+    },
+    {
+      "name": "set_api_limits",
+      "description": "Update API rate limits and usage quotas"
+    },
+    {
+      "name": "get_api_limits",
+      "description": "Get current API rate limits and usage"
+    }
+  ]
+}
\ No newline at end of file