SQLite MCP Server

Last Updated October 9, 2025 - Production/Stable v2.6.3

Enterprise-grade SQLite with AI-native JSON operations & intelligent workflow automation – v2.6.3

Transform SQLite into a powerful, AI-ready database engine with 73 specialized tools for advanced analytics, JSON operations, text processing, vector search, geospatial operations, and intelligent workflow automation.

---

📋 Table of Contents

Quick Start

• ✅ Quick Test - Verify Everything Works
• 🚀 Quick Start
• 🔥 Core Capabilities
• 🏢 Enterprise Features

Configuration & Usage

• 📚 MCP Client Configuration
• 🎨 Usage Examples
• 📊 Tool Categories

Resources & Information

• 🏆 Why Choose SQLite MCP Server?
• 🔍 AI-Powered Wiki Search
• 📚 Complete Documentation
• 🔗 Additional Resources
• 🚀 Quick Links
• 📈 Project Stats

---

✅ Quick Test - Verify Everything Works

Test all 73 tools in 30 seconds!

Quick smoke test:

python test_runner.py --quick

Standard comprehensive test (recommended):

python test_runner.py --standard

Full test suite with edge cases:

python test_runner.py --full

Expected output:

🚀 SQLite MCP Server Comprehensive Test Suite v2.6.3
================================================================

🔍 Environment Detection:
  ✅ SQLite 3.50.4 (JSONB supported)
  ✅ Python 3.12.11  
  ✅ MCP 1.14.0
  ✅ Pyright strict type checking: PASS

📊 Testing 73 Tools across 14 categories...

✅ Core Database Operations (8/8 passed)
✅ JSON Helper Tools (6/6 passed)  
✅ JSON Operations (12/12 passed)  
✅ Text Processing (8/8 passed)
🎉 SUCCESS: All 73 tools tested successfully!

🛡️ Security Testing

NEW: Comprehensive SQL injection protection testing

Test SQL injection protection (from tests directory):

cd tests && python test_sql_injection.py

Expected result: 🛡️ Overall security posture: STRONG

What it tests:

• Protection against the SQL injection vulnerability found in original Anthropic SQLite MCP server
• 11 different attack vectors including multiple statements, UNION injection, blind injection
• Parameter binding protection with malicious payloads
• Stacked queries and comment-based injection attempts

⬆️ Back to Table of Contents

🚀 Quick Start

Option 1: Docker (Recommended)

Pull and run instantly:

docker pull writenotenow/sqlite-mcp-server:latest

Run with volume mount:

docker run -i --rm \
  -v $(pwd):/workspace \
  writenotenow/sqlite-mcp-server:latest \
  --db-path /workspace/database.db

🛡️ Supply Chain Security

For enhanced security and reproducible builds, use SHA-pinned images:

Find available SHA tags at: https://hub.docker.com/r/writenotenow/sqlite-mcp-server/tags Look for tags starting with "master-" or "sha256-" for cryptographically verified builds

Option 1: Human-readable timestamped builds (recommended)

docker pull writenotenow/sqlite-mcp-server:master-YYYYMMDD-HHMMSS-<commit>

Option 2: Multi-arch manifest digest (maximum security)

docker pull writenotenow/sqlite-mcp-server@sha256:<manifest-digest>

Example: Run with cryptographically verified image

docker run -i --rm \
  -v $(pwd):/workspace \
  writenotenow/sqlite-mcp-server:master-YYYYMMDD-HHMMSS-<commit> \
  --db-path /workspace/database.db

How to Find SHA Tags:

1. Visit Docker Hub Tags
2. For convenience: Use master-YYYYMMDD-HHMMSS-<commit> tags (human-readable, multi-arch)
3. For maximum security: Use sha256-<hash> tags (manifest digests, immutable)

Understanding SHA Tags:

• 🏷️ master-YYYYMMDD-HHMMSS-<commit> - Human-readable, timestamped, multi-arch safe
• 🔒 sha256-<manifest-digest> - Multi-arch manifest digest (works on all architectures)
• ⚠️ Architecture-specific digests - Only for debugging specific architectures

Security Features:

• ✅ Build Provenance - Cryptographic proof of build process
• ✅ SBOM Available - Complete software bill of materials
• ✅ Supply Chain Attestations - Verifiable build integrity
• ✅ Reproducible Builds - Exact image verification for compliance

Option 2: Python Installation

Install from PyPI:

pip install sqlite-mcp-server-enhanced

Or install from source:

git clone https://github.com/neverinfamous/sqlite-mcp-server.git

Navigate to directory:

cd sqlite-mcp-server

Install requirements:

pip install -r requirements.txt

Run the server:

python start_sqlite_mcp.py --db-path ./database.db

Option 3: Test in 30 Seconds

Clone repository:

git clone https://github.com/neverinfamous/sqlite-mcp-server.git

Navigate to directory:

cd sqlite-mcp-server

Run quick test:

python test_runner.py --quick

🆕 NEW in v2.6.0: Complete JSON Operations Suite

5 Major Improvements in this release:

• 🎯 JSON Helper Tools - 6 specialized tools for simplified JSON operations with path validation and merging
• 🤖 JSON Auto-Normalization - Automatically fixes Python-style JSON with configurable strict mode
• 🛡️ Parameter Binding Interface - Enhanced MCP tools with SQL injection prevention
• 📦 Automatic Parameter Serialization - Direct object/array parameters, no manual JSON.stringify()
• 🧠 Enhanced JSON Error Diagnostics - Intelligent error categorization with contextual guidance

🔥 Core Capabilities

• 📊 Statistical Analysis - Descriptive stats, percentiles, time series analysis
• 🔍 Advanced Text Processing - Regex, fuzzy matching, phonetic search, similarity
• 🧠 Vector/Semantic Search - AI-native embeddings, cosine similarity, hybrid search
• 🗺️ SpatiaLite Geospatial - Enterprise GIS with spatial indexing and operations
• 🔐 Transaction Safety - Auto-wrapped transactions with rollback protection
• 🎛️ 73 Specialized Tools - Complete database administration and analytics suite

🏢 Enterprise Features

• 📈 Business Intelligence - Integrated insights memo and workflow automation
• 🔄 Backup/Restore - Enterprise-grade operations with integrity verification
• 🎯 Full-Text Search (FTS5) - Advanced search with BM25 ranking and snippets
• 🏗️ Virtual Tables - Smart CSV/JSON import with automatic type inference
• ⚙️ Advanced PRAGMA - Complete SQLite configuration and optimization

⬆️ Back to Table of Contents

📚 MCP Client Configuration

Claude Desktop

{
  "mcpServers": {
    "sqlite-mcp-server": {
      "command": "python",
      "args": ["/path/to/sqlite-mcp-server/start_sqlite_mcp.py", "--db-path", "/path/to/database.db"]
    }
  }
}

Docker with Claude Desktop

{
  "mcpServers": {
    "sqlite-mcp-server": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "-v", "/path/to/project:/workspace", "writenotenow/sqlite-mcp-server:latest", "--db-path", "/workspace/database.db"]
    }
  }
}

⬆️ Back to Table of Contents

🎨 Usage Examples

Data Analysis Workflow

1. Quick validation:

python test_runner.py --quick

2. Start with your data:

python start_sqlite_mcp.py --db-path ./sales_data.db

3. Use with Claude/Cursor for:
   • Statistical analysis of your datasets
   • Text processing and pattern extraction
   • Vector similarity search
   • Geospatial analysis and mapping
   • Business intelligence insights

Docker Development

Development with live reload:

docker run -i --rm \
  -v $(pwd):/workspace \
  -e SQLITE_DEBUG=true \
  writenotenow/sqlite-mcp-server:latest \
  --db-path /workspace/dev.db

🎯 JSON Helper Tools - Simplified JSON Operations

NEW in v2.6.0: Six powerful JSON helper tools that make complex JSON operations simple:

// ✅ Insert JSON with auto-normalization
json_insert({
  "table": "products",
  "column": "metadata", 
  "data": {'name': 'Product', 'active': True, 'price': None}
})

// ✅ Update JSON by path
json_update({
  "table": "products",
  "column": "metadata",
  "path": "$.price",
  "value": 29.99,
  "where_clause": "id = 1"
})

// ✅ Query JSON with complex filtering
json_query({
  "table": "products",
  "column": "metadata",
  "filter_paths": {"$.category": "electronics"},
  "select_paths": ["$.name", "$.price"]
})

JSON Helper Tools:

• 🎯 json_insert - Insert JSON data with auto-normalization
• 🔄 json_update - Update JSON by path with creation support
• 🔍 json_select - Extract JSON data with multiple output formats
• 🔎 json_query - Complex JSON filtering and aggregation
• ✅ json_validate_path - Validate JSON paths with security checks
• 🔗 json_merge - Merge JSON objects with conflict resolution

Auto-normalization still works:

• 🔧 Single quotes → Double quotes
• 🔧 Python True/False → JSON true/false
• 🔧 Python None → JSON null
• 🔧 Trailing commas removed
• 🛡️ Security validation prevents malicious input

🧠 Enhanced JSON Error Diagnostics

Enhanced in v2.6.0: When JSON validation fails, get intelligent, contextual error messages with specific guidance:

// ❌ Invalid JSON input:
validate_json('{key_without_quotes: "value"}')

// ✅ Enhanced error response:
{
  "valid": false,
  "error": "Expecting property name enclosed in double quotes: line 1 column 2 (char 1)",
  "enhanced_message": "JSON validation failed (structural_syntax): Expecting property name...",
  "error_context": {
    "error_type": "structural_syntax",
    "security_concern": false,
    "suggestions": [
      "Ensure all object keys are properly quoted strings",
      "Check for missing colons (:) between keys and values",
      "Verify proper key-value pair structure: \"key\": \"value\""
    ]
  }
}

Enhanced Error Categories:

• 🔧 Structural Issues - Missing quotes, colons, brackets with specific fix suggestions
• 🛡️ Security Warnings - Detects potential SQL injection patterns in JSON strings
• 📝 Encoding Problems - Character encoding and escape sequence guidance
• 🎯 Context-Aware Tips - Line/column position with targeted recommendations

🛡️ Enhanced Parameter Binding + Auto-Serialization

NEW in v2.6.0: Built-in SQL injection protection with automatic JSON serialization:

// ✅ SECURE: Parameter binding prevents injection
read_query({
  "query": "SELECT * FROM users WHERE name = ? AND age > ?",
  "params": ["John", 25]
})

// ✅ NEW: Direct object/array parameters (auto-serialized)
write_query({
  "query": "INSERT INTO products (metadata, tags) VALUES (?, ?)",
  "params": [
    {"name": "Product", "price": 29.99, "active": true},  // Auto-serialized to JSON
    ["electronics", "featured", "new"]                    // Auto-serialized to JSON
  ]
})

// ✅ SIMPLIFIED: No more manual JSON.stringify()
// Before v2.6.0:
write_query({
  "query": "INSERT INTO table (data) VALUES (?)",
  "params": [JSON.stringify({"key": "value"})]  // Manual serialization required
})

// After v2.6.0:
write_query({
  "query": "INSERT INTO table (data) VALUES (?)",
  "params": [{"key": "value"}]  // Automatic serialization!
})

v2.6.0 Benefits:

• 🛡️ SQL Injection Prevention - Parameter binding treats malicious input as literal data
• 📦 Auto-Serialization - Objects and arrays automatically converted to JSON strings
• 🔄 Backward Compatible - Existing queries continue to work unchanged
• ⚡ Better Performance - Query plan caching and parameter optimization
• 📝 Cleaner API - No manual JSON.stringify() or parameter preparation needed

⬆️ Back to Table of Contents

📊 Tool Categories

The SQLite MCP Server provides 73 specialized tools across 14 categories:

💡 Want the complete tool list? See the detailed tool reference with descriptions for all 73 tools, 7 resources, and 7 prompts.

Category	Tool Count	Description
Core Database	15	CRUD operations, schema management, transactions
JSON Helper Tools	6	Simplified JSON operations, path validation, merging
Text Processing	9	Regex, fuzzy matching, phonetic search, similarity
Statistical Analysis	8	Descriptive stats, percentiles, time series
Virtual Tables	8	CSV, R-Tree, series generation
Semantic Search	8	Embeddings, similarity, hybrid search
Geospatial	7	Spatial indexing, geometric operations
PRAGMA Operations	5	Configuration, optimization, introspection
Full-Text Search	3	FTS5 creation, indexing, BM25 ranking
Vector Optimization	2	ANN search, clustering, performance
Data Analysis	2	Smart CSV/JSON import with type inference
Resources	7	Database meta-awareness, performance insights
Prompts	7	Guided workflows, optimization recipes

💡 Cursor Users: You can enable only the categories you need in your MCP client settings to reduce tool noise and improve stability. Each number above shows the count of tools in that category.

⬆️ Back to Table of Contents

🏆 Why Choose SQLite MCP Server?

✅ AI-Friendly - JSON auto-normalization and intelligent error messages reduce debugging time
✅ Just Works - Built-in security and parameter binding with zero configuration
✅ Smart Diagnostics - Enhanced error context provides actionable guidance when issues occur
✅ Type Safe - Passes strict Pyright type checking in Cursor for maximum code quality
✅ Instantly Testable - Validate all 73 tools in 30 seconds
✅ Production Ready - Enterprise-grade testing and validation
✅ Comprehensive - Everything you need in one package
✅ Docker Ready - Containerized for easy deployment
✅ Zero Breaking Changes - All existing code continues to work

⬆️ Back to Table of Contents

🔍 AI-Powered Wiki Search

→ Search the Documentation with AI

Can't find what you're looking for? Use our AI-powered search interface to search both SQLite and PostgreSQL MCP Server documentation:

• 🤖 Natural Language Queries - Ask questions in plain English and get AI-generated answers
• ⚡ Instant Results - AI-enhanced answers with source attribution from both wikis
• 📚 Comprehensive Coverage - Searches all 73 SQLite tools + 63 PostgreSQL tools (136 total)
• 🎯 Smart Context - Understands technical questions and provides relevant examples
• 🔄 Dual Search Modes - AI-Enhanced for synthesized answers, or Raw Docs for direct chunks

Example queries:

• "How do I prevent SQL injection attacks?"
• "What statistical analysis tools are available?"
• "How do I set up vector search with embeddings?"
• "How do I use JSON helper tools for data normalization?"
• "What SpatiaLite geospatial operations are available?"

→ Try AI Search Now

The search interface uses Cloudflare's AI Search technology to provide intelligent, context-aware answers from comprehensive wiki documentation for both SQLite and PostgreSQL MCP Servers.

⬆️ Back to Table of Contents

📚 Complete Documentation

→ Wiki: Comprehensive Documentation & Examples

Comprehensive documentation including:

• Detailed tool reference - All 73 tools with examples
• Advanced configuration - Performance tuning and optimization
• Integration guides - MCP clients, Docker, CI/CD
• Feature deep-dives - Text processing, vector search, geospatial
• Best practices - Query patterns, troubleshooting, workflows
• API reference - Complete tool schemas and parameters

→ GitHub Gists: Practical Examples & Use Cases

9 curated gists with real-world examples:

• JSON Helper Tools - Simplified JSON operations with auto-normalization
• Vector/Semantic Search - AI-native embeddings and similarity search
• SpatiaLite GIS - Geospatial operations and spatial indexing
• Performance Optimization - Query tuning and index recommendations
• Security Best Practices - SQL injection prevention and safe queries
• Real-World Use Cases - Business intelligence and data analysis workflows
• Database Migration - Schema evolution and data transformation
• Docker Deployment - Production containerization strategies
• Complete Feature Showcase - All 73 tools with comprehensive examples

⬆️ Back to Table of Contents

🔗 Additional Resources

• Testing Guide - Comprehensive testing documentation
• Contributing - How to contribute to the project
• Security Policy - Security guidelines and reporting
• Code of Conduct - Community guidelines
• Docker Hub - Container images
• GitHub Releases - Version history
• Adamic Support Blog - Project announcements and releases

⬆️ Back to Table of Contents

🚀 Quick Links

Action	Command
AI-Powered Search	search.adamic.tech
Test Everything	python test_runner.py --standard
Docker Quick Start	docker run -i --rm -v $(pwd):/workspace writenotenow/sqlite-mcp-server:latest
Install from PyPI	pip install sqlite-mcp-server-enhanced
View Full Docs	docs/sqlite-mcp-server.wiki
Report Issues	GitHub Issues

⬆️ Back to Table of Contents

📈 Project Stats

• 73 Tools across 14 categories (all tested and verified ✅)
• 2,000+ lines of comprehensive documentation
• Multi-platform support (Windows, Linux, macOS)
• Docker images for amd64 and arm64
• Strict type checking with Pyright for code quality
• Enterprise testing with comprehensive validation
• Active development with regular updates

⬆️ Back to Table of Contents