AWS Compliance MCP Server
Enables LLMs to perform automated compliance scanning of AWS infrastructure against PCI-DSS, CIS, and Well-Architected frameworks. It provides tools for scanning, remediation, and audit report generation.
README
AWS Compliance MCP Server
A Model Context Protocol (MCP) server that connects an LLM (Claude, GPT-4, etc.) to real AWS infrastructure and performs automated compliance scanning against:
- PCI-DSS v3.2.1 (18 controls across 8 requirements)
- CIS AWS Foundations Benchmark v1.4 (19 controls across 5 sections)
- AWS Well-Architected Framework (14 checks across all 5 pillars)
Architecture
Claude / Any MCP Client
│ MCP Protocol (stdio)
▼
aws-compliance-mcp (FastMCP server)
├── ConfigTools → AWS Config APIs
├── SecurityHubTools → Security Hub APIs
└── ComplianceScanner → Rule mappings + scoring
Prerequisites
- Python 3.10+
- AWS credentials with read access to Config and Security Hub
- AWS Config and Security Hub enabled in your target region
Installation
# Clone the repo
git clone https://github.com/YOUR_USERNAME/aws-compliance-mcp.git
cd aws-compliance-mcp
# Create virtual environment
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
# Install the package
pip install -e ".[dev]"
# Or with pip from requirements
pip install -r requirements.txt
pip install -r requirements-dev.txt
Configuration
cp .env.example .env
# Edit .env with your AWS region and optional profile
.env.example:
AWS_REGION=us-east-1
# AWS_PROFILE=my-profile # Uncomment to use a named profile
Running the Server
# Run directly
python -m aws_compliance_mcp.server
# Or via installed script
aws-compliance-mcp
Connect to Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS):
{
"mcpServers": {
"aws-compliance": {
"command": "/path/to/.venv/bin/python",
"args": ["-m", "aws_compliance_mcp.server"],
"env": {
"AWS_REGION": "us-east-1",
"AWS_PROFILE": "your-profile"
}
}
}
}
Available MCP Tools
| Tool | Description |
|---|---|
scan_pcidss_compliance |
Full PCI-DSS v3.2.1 scan |
scan_cis_compliance |
CIS Benchmark v1.4 scan |
scan_well_architected |
Well-Architected scan (all pillars or single) |
scan_all_frameworks |
Unified scan with overall risk score |
get_remediation_steps |
Step-by-step fix for a specific control |
generate_audit_report |
Audit-ready report with top-10 priorities |
get_config_compliance_summary |
Overall Config rule compliance % |
list_config_rules |
All Config rules + compliance state |
get_config_rule_status |
Single rule details + failing resources |
get_noncompliant_resources |
NON_COMPLIANT resources |
get_security_hub_summary |
Finding counts by severity |
get_security_hub_findings |
Filtered findings list |
get_security_hub_score |
Pass/fail ratio per standard |
Example Claude prompts
"Run a full PCI-DSS compliance scan in us-east-1 and tell me what's failing."
"What are my top 10 remediation priorities across all frameworks?"
"Give me step-by-step remediation for PCI.10.1"
"What's my Security Hub score and how many CRITICAL findings do I have?"
MCP Resources
| URI | Description |
|---|---|
compliance://pcidss/controls |
Full PCI-DSS control catalog |
compliance://cis/controls |
Full CIS control catalog |
compliance://well-architected/pillars |
Well-Architected pillar checks |
Running Tests
# All tests
make test
# With coverage
make coverage
# Lint
make lint
# Or directly
pytest tests/ -v
Tests use moto to mock all AWS API calls — no real AWS account needed.
Project Structure
aws-compliance-mcp/
├── src/aws_compliance_mcp/
│ ├── server.py # FastMCP server + all tool/resource definitions
│ ├── aws_client.py # Shared boto3 client with retry config
│ ├── tools/
│ │ ├── config_tools.py # AWS Config API wrapper
│ │ ├── securityhub_tools.py # Security Hub API wrapper
│ │ └── compliance_scanner.py # Framework scanning + reporting
│ └── rules/
│ ├── pcidss.py # PCI-DSS v3.2.1 control mappings
│ ├── cis.py # CIS Benchmark v1.4 control mappings
│ └── well_architected.py # Well-Architected checks
├── tests/
│ ├── conftest.py
│ ├── test_config_tools.py
│ ├── test_securityhub_tools.py
│ └── test_compliance_scanner.py
├── pyproject.toml
├── Makefile
└── .env.example
Required IAM Permissions
{
"Version": "2012-10-17",
"Statement": [
{
"Effect": "Allow",
"Action": [
"config:DescribeComplianceByConfigRule",
"config:GetComplianceSummaryByConfigRule",
"config:GetComplianceDetailsByConfigRule",
"config:DescribeComplianceByResource",
"config:DescribeConfigRules",
"securityhub:GetFindings",
"securityhub:GetEnabledStandards",
"securityhub:DescribeStandardsControls"
],
"Resource": "*"
}
]
}
CI/CD
GitHub Actions runs tests on Python 3.10 and 3.12 on every push. See .github/workflows/ci.yml.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.