Enterprise Secure AWS MCP Server
Enterprise Secure AWS Model Context Protocol (MCP) Server
Produktionsreifer Model Context Protocol (MCP) Server, bereitgestellt auf AWS Lambda & API Gateway, mit 100 % Terraform IaC, Zero-Trust-IAM RBAC und Abwehrmaßnahmen gemäß der OWASP-LLM-Top-10.
AWS-Architekturübersicht

Related MCP server: AWS Sage
Funktionsumfang
Standard-MCP-Spezifikationen: Volle Unterstützung für Tool Discovery (
/mcp/tools) und Tool Execution (/mcp/tools/call) des Anthropic Model Context Protocol (MCP).IAM-Rollenbasierte Zugriffskontrolle (RBAC): Beschränkt die Tool-Ausführung anhand der Rolle (
sre_read_only,security_auditor,sre_admin).OWASP-LLM02-Abwehr: Edge-Säuberung, die Subshell-Metazeichen-Injektionen (
[;&|'$]) blockiert.OWASP-LLM06-Abwehr: Automatisches Schwärzen der Ausgabepuffer, die AWS-Zugriffsschlüssel (
AKIA...) und personenbezogene Daten (PII) entfernt.Interaktive Web-UI-Playground: Dark-Mode-Schnittstelle, die auf
http://localhost:8080läuft (make ui).Keine Cloud-Kosten im Leerlauf: Provisioniert über Terraform mittels Ein-Befehl-Deployment (
make deploy) und sofortigem Abbau (make destroy).
Web-UI-Plattform Demo
Nachfolgend sehen Sie die automatisierte High-Definition-Interactive-Demonstration, die die RBAC-Zugriffsverweigerung, die Abwehr von Befehlsinjektionen und die erfolgreiche Ausführung des Server-Neustarts unter der Rolle sre_admin zeigt:

Voraussetzungen & Systems-
Einrichtung
Stellen Sie vor dem Einrichten des Projekts sicher, dass in Ihrer Umgebung Python 3.11+, Terraform und die AWS CLI installiert sind.
1. Systemabhängigkeiten (Linux / Ubuntu / Debian)
sudo apt update && sudo apt install python3-venv python3-pip terraform awscli -y2. Umgebungseinrichtung
Klonen Sie das Repository und installieren Sie die Projektabhängigkeiten in einer isolierten virtuellen Umgebung (.venv):
git clone https://github.com/patelketul1230/aws-mcp-enterprise-server.git
cd aws-mcp-enterprise-server
# Create .venv and install dependencies
make installRepository-Struktur
aws-mcp-enterprise-server/
├── docs/
│ ├── assets/ # Centralized PNG images & HD animations
│ │ ├── mcp_aws_figma_architecture_diagram.png
│ │ ├── mcp_ui_testing_demo.webp
│ │ ├── mcp_ui_testing_demo.gif
│ │ ├── mcp_ui_playground_success_screenshot.png
│ │ ├── mcp_ui_playground_admin_success_screenshot.png
│ │ └── mcp_ui_playground_blocked_screenshot.png
│ ├── blogs/ # 3 Medium Sub-Blogs ready for publication
│ │ ├── blog_1_1_mcp_decoded.md
│ │ ├── blog_1_2_mcp_rbac_python.md
│ │ └── blog_1_3_mcp_tool_poisoning.md
│ ├── architecture.md # Figma diagrams & sequence flows
│ ├── 5w_and_how.md # 5 Ws + 1 H problem statement & threat model
│ └── demo_execution.log # Verified test execution log
├── src/
│ ├── clients/ # AWS Bedrock Converse API integration client
│ ├── middleware/ # MCPSecurityValidator (RBAC & DLP)
│ ├── tools/ # CloudWatch & Terraform tool modules
│ ├── ui/ # Interactive Web UI Playground (index.html)
│ └── server.py # AWS Lambda & FastAPI server handler
├── terraform/ # 100% Terraform IaC (API Gateway, IAM, Lambda)
├── tests/ # Pytest suite & live verification scripts
├── Makefile # Lifecycle hooks (make deploy / make destroy / make ui)
└── README.md # Project documentationSchnellstart-Anleitung
1. Abhängigkeiten installieren
make install2. Tests ausführen
make test3. Infrastruktur in AWS bereitstellen
make deployAusgaben:
API-Endpunkt: Dynamische, von Terraform generierte Ausgabe
CloudWatch Logging Group:
/aws/mcp/aws-mcp-enterprise-server-dev
4. Interaktiven Web-UI-Playground starten
make uiÖffnen Sie http://localhost:8080 in Ihrem Browser, um Tool-Aufrufe, RBAC-Rollen und Injektionsabwehrmaßnahmen visuell zu testen!
5. Testen über das Terminal (curl)
# Health Check
curl -s <YOUR_API_ENDPOINT>/health
# Tool Discovery Catalog
curl -s <YOUR_API_ENDPOINT>/mcp/tools
# Valid Tool Call (CloudWatch Query)
curl -s -X POST <YOUR_API_ENDPOINT>/mcp/tools/call \
-H "Content-Type: application/json" \
-H "X-User-Role: sre_read_only" \
-d '{"tool_name": "query_cloudwatch_logs", "arguments": {"log_group": "/aws/lambda/payment-dev", "filter_pattern": "ERROR"}}'
# Command Injection Attack (Returns 403 Forbidden)
curl -s -X POST <YOUR_API_ENDPOINT>/mcp/tools/call \
-H "Content-Type: application/json" \
-H "X-User-Role: sre_read_only" \
-d '{"tool_name": "query_cloudwatch_logs", "arguments": {"log_group": "/aws/lambda/payment-dev; rm -rf /", "filter_pattern": "ERROR"}}'6. AWS-Ressourcen abbauen
make destroyLinks zur technischen Dokumentation
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseBqualityDmaintenanceEnables AI models to interact with AWS Lambda functions via the MCP protocol, allowing access to private resources, real-time data, and custom computation in a secure environment.2109MIT
- AlicenseNot gradedqualityCmaintenanceA unified MCP server for AWS that enables natural language infrastructure management, cross-service resource discovery, and dependency mapping. It features 30 intelligent tools for cost optimization, incident investigation, and multi-account operations protected by a robust safety system.5MIT
- AlicenseCqualityDmaintenanceEnables secure, zero-trust access to MCP tools through short-lived, signed capability leases that bind tool execution to specific sessions, intents, and constraints. Prevents prompt injection attacks and privilege escalation with dynamic risk scoring, policy enforcement, and tamper-evident audit logging.41MIT
- AlicenseNot gradedqualityCmaintenanceA read-only MCP server for safe, structured investigation of AWS serverless resources, providing curated tools for tracing dependencies, permissions, and failures without exposing raw SDK access.MIT
Related MCP Connectors
Remote MCP for Copilot CLI switch gate MCP, structured receipts, audit logs, and reviewer-ready evid
A paid remote MCP for AI SDK eval dashboard, built to return verdicts, receipts, usage logs, and aud
A paid remote MCP for ClawManager, built to return verdicts, receipts, usage logs, and audit-ready J
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/patelketul1230/aws-mcp-enterprise-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server