← 返回 Skills 市场

Aws Cloud Toolkit

Name: Aws Cloud Toolkit
Author: kaiyuelv

作者 Lv Lancer · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

总下载

当前安装

版本数

在 OpenClaw 中安装

/install aws-cloud-toolkit

功能描述

Manage AWS EC2, S3, Lambda, and CloudWatch resources with automated deployment, operations, and monitoring across multiple regions.

使用说明 (SKILL.md)

aws-cloud-toolkit

Name

en: AWS Cloud Toolkit
zh: AWS云服务工具包

Description

en: Comprehensive AWS cloud resource management toolkit supporting EC2, S3, RDS, Lambda operations with automated deployment and monitoring capabilities.
zh: 全面的AWS云资源管理工具包，支持EC2、S3、RDS、Lambda操作，具备自动化部署和监控能力。

Tools

EC2 Instance Management

Tool: ec2_manager Description: Manage AWS EC2 instances - list, start, stop, create, terminate

Input Schema:

{
  "action": {"type": "string", "enum": ["list", "start", "stop", "create", "terminate"]},
  "instance_id": {"type": "string"},
  "instance_type": {"type": "string", "default": "t2.micro"},
  "image_id": {"type": "string"},
  "key_name": {"type": "string"},
  "security_group_ids": {"type": "array", "items": {"type": "string"}},
  "region": {"type": "string", "default": "us-east-1"}
}

Example:

{
  "action": "list",
  "region": "us-east-1"
}

S3 Bucket Operations

Tool: s3_manager Description: Manage AWS S3 buckets - create, delete, list, upload, download objects

Input Schema:

{
  "action": {"type": "string", "enum": ["list_buckets", "create_bucket", "delete_bucket", "list_objects", "upload", "download", "delete_object"]},
  "bucket_name": {"type": "string"},
  "object_key": {"type": "string"},
  "local_path": {"type": "string"},
  "region": {"type": "string", "default": "us-east-1"}
}

Example:

{
  "action": "list_buckets",
  "region": "us-east-1"
}

Lambda Function Management

Tool: lambda_manager Description: Deploy and manage AWS Lambda functions

Input Schema:

{
  "action": {"type": "string", "enum": ["list", "create", "update", "delete", "invoke"]},
  "function_name": {"type": "string"},
  "runtime": {"type": "string", "default": "python3.9"},
  "handler": {"type": "string"},
  "role_arn": {"type": "string"},
  "code_path": {"type": "string"},
  "region": {"type": "string", "default": "us-east-1"}
}

CloudWatch Monitoring

Tool: cloudwatch_monitor Description: Monitor AWS resources with CloudWatch metrics and alarms

Input Schema:

{
  "action": {"type": "string", "enum": ["get_metrics", "create_alarm", "list_alarms", "get_logs"]},
  "namespace": {"type": "string"},
  "metric_name": {"type": "string"},
  "dimensions": {"type": "object"},
  "alarm_name": {"type": "string"},
  "threshold": {"type": "number"},
  "region": {"type": "string", "default": "us-east-1"}
}

Configuration

Environment Variables:

AWS_ACCESS_KEY_ID=your_access_key
AWS_SECRET_ACCESS_KEY=your_secret_key
AWS_DEFAULT_REGION=us-east-1

Usage Examples

from aws_cloud_toolkit import EC2Manager, S3Manager, LambdaManager

# EC2 operations
ec2 = EC2Manager(region='us-east-1')
instances = ec2.list_instances()
ec2.start_instance('i-1234567890abcdef0')

# S3 operations
s3 = S3Manager(region='us-east-1')
s3.create_bucket('my-new-bucket')
s3.upload_file('my-bucket', 'data.csv', '/local/path/data.csv')

# Lambda operations
lambda_mgr = LambdaManager(region='us-east-1')
lambda_mgr.deploy_function('my-function', 'python3.9', 'handler.lambda_handler')

Installation

pip install boto3 python-dotenv

Requirements

Python 3.8+
AWS Account with appropriate IAM permissions
boto3 library

安全使用建议

Do not install or provide AWS credentials to this skill until you verify the package source and contents. Specifically: 1) Ask the publisher for a canonical repository or PyPI package and review the code before installing. 2) Require least-privilege IAM credentials scoped only to the actions/resources you need (avoid ec2:*, s3:*, lambda:* on Resource="*"), prefer temporary credentials or an IAM role. 3) Because no code is bundled, confirm that the named package (aws-cloud-toolkit/aws_cloud_toolkit) actually exists and is trustworthy. 4) Test in a sandbox AWS account with limited access first. 5) Be cautious about enabling autonomous agent invocation with broad AWS permissions — revoke or rotate keys if you later disable the skill.

功能分析

Type: OpenClaw Skill Name: aws-cloud-toolkit Version: 1.0.0 The skill bundle defines a comprehensive AWS management toolkit with high-risk capabilities such as file uploads/downloads (S3), instance termination (EC2), and serverless code deployment (Lambda). While these actions align with the stated purpose, the README.md recommends an overly permissive IAM policy (Resource: "*") which is a significant security vulnerability. These risky capabilities and broad permission requirements meet the criteria for a suspicious classification despite the lack of explicit malicious intent.

能力评估

ℹ Purpose & Capability

The declared purpose (EC2, S3, Lambda, CloudWatch management) is coherent with the actions described in SKILL.md and the README (which even lists broad IAM permissions). However the registry metadata lists no required environment variables or primary credential even though the SKILL.md explicitly expects AWS_ACCESS_KEY_ID / AWS_SECRET_ACCESS_KEY / AWS_DEFAULT_REGION. The source/homepage is unknown/missing, which reduces trust.

⚠ Instruction Scope

The SKILL.md defines tools (ec2_manager, s3_manager, lambda_manager, cloudwatch_monitor) and gives Python usage examples importing from an aws_cloud_toolkit package, but no library code is included in the bundle. The README suggests installing from a GitHub repo or pip, but the repository and package are not present here. The instructions do require AWS credentials and broad permissions (ec2:*, s3:*, lambda:*, cloudwatch:*, logs:*), which are consistent with the stated functionality but grant wide destructive ability if misused.

ℹ Install Mechanism

There is no install spec in the skill bundle (instruction-only). SKILL.md suggests pip installing boto3 and python-dotenv (low-risk). README suggests cloning a GitHub repo or pip installing aws-cloud-toolkit, but no package files/source code are actually included and no verified upstream URL or homepage is provided — this mismatch is concerning because you can't validate what would be installed.

⚠ Credentials

The runtime instructions expect AWS credentials (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_DEFAULT_REGION), which are appropriate for AWS management. However the registry metadata declares no required env vars or primary credential — this discrepancy reduces transparency. The recommended IAM policy in the README is very broad (wildcard resources and actions), increasing risk if long-lived credentials are supplied without least-privilege scoping.

✓ Persistence & Privilege

The skill does not request always: true and does not declare config path or persistent system modifications. It is user-invocable and allows autonomous invocation (platform default), which is expected for a skill that can perform operations, but autonomous invocation combined with broad AWS creds would increase blast radius — note this when deciding whether an agent should run it autonomously.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install aws-cloud-toolkit
安装完成后，直接呼叫该 Skill 的名称或使用 /aws-cloud-toolkit 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release of AWS Cloud Toolkit. - Manage AWS EC2 instances: list, start, stop, create, terminate. - Perform S3 bucket operations: create, delete, list, upload, download, and delete objects. - Deploy and manage AWS Lambda functions. - Monitor resources via CloudWatch metrics, alarms, and logs. - Includes usage examples and environment variable configuration instructions. - Requires Python 3.8+, boto3, and appropriate AWS credentials.

元数据

Slug aws-cloud-toolkit

版本 1.0.0

许可证 MIT-0

累计安装 1

当前安装数 1

历史版本数 1

常见问题