LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|
Toolofficial publisher

Outils cloud et Kubernetes (AWS, GCP, Azure, Terraform)

Tool : A precise action the AI can call · requires divers

Control infrastructure from an agent.

The largest possible blast radius: read-only by default, writes under review.

Docs
Overall score
5.7 /10
Security: serious reservations
Utility
7.0
Reliability
6.0
Security
Serious reservations
Ease
Technical · Technical

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Managing cloud infrastructure from an AI agent requires connecting the assistant to AWS, GCP, Azure, or Terraform services without losing control over executed actions. These open source MCP servers, published by AWS Labs, address this need by exposing cloud capabilities through the MCP protocol.

In practice, the agent has tools to query AWS documentation in real time, manage Kubernetes clusters on Amazon EKS, deploy containers via Amazon ECS, manage resources with the Cloud Control API, and orchestrate Infrastructure as Code workflows with CDK and Terraform. Each MCP server connects to your AI client (Cursor, VS Code, Claude Desktop, Cline, Windsurf) via stdio.

These tools are aimed at developers and cloud engineers who want to delegate repetitive infrastructure tasks to an agent while maintaining full traceability via CloudTrail. The AWS-hosted version adds syntactic validation of API calls and IAM-based permission management with zero credential exposure.

This is not suitable for users without cloud experience: write and execution actions on the infrastructure are real and can incur costs or delete resources. Human review of actions proposed by the agent remains essential.

What it offers

  • ·Real-time access to official AWS documentation, API references, and getting started guides.
  • ·Kubernetes cluster management via the Amazon EKS MCP server, including application deployment.
  • ·Container orchestration with Amazon ECS MCP Server for deployment workflows.
  • ·Infrastructure as Code with the AWS IaC MCP server: CloudFormation, CDK, security validation, and deployment troubleshooting.
  • ·Audited AWS interactions via CloudTrail with syntactic API call validation and IAM permissions, available on the hosted AWS MCP server (in preview).
  • ·Execution of Lambda functions as AI tools to access private resources via AWS Lambda Tool MCP Server.
  • ·Operations on relational and NoSQL databases: DynamoDB, Aurora PostgreSQL, Aurora MySQL, DocumentDB, Neptune, Redshift, and others.

In practice

Price
Free
Commercial use
Commercial use: yes
Available on
mcp, sdk
Host
divers
FR interface
No

Security and data

Rating
Security: serious reservations
Capabilities
readswritesexecutesnetwork
Your data
Data with third parties
Maintenance
Active

Links and repository checked on 25 October 2026 · added on 1 April 2025

How to install Outils cloud et Kubernetes (AWS, GCP, Azure, Terraform)

1. Make sure you have a compatible MCP client (Cursor, VS Code, Claude Desktop, Cline, Windsurf). 2. Ensure your AWS credentials are configured with IAM permissions limited to what is strictly necessary. 3. Choose the MCP server(s) suited to your use case from the awslabs/mcp repository list. 4. Add the chosen server configuration to your client's MCP settings file (e.g. .cursor/mcp.json, .vscode/mcp.json). 5. Use stdio transport only; do not configure SSE, as this mode is no longer supported. 6. Test first with a read-only server (documentation) before enabling servers with write access to the infrastructure.
MCP AWS, kubectl, Terraform

Official documentation ↗

In the same category

Whole category Automation & agents →