de en es ja ko ru zh

S3 MCP Server

Name: S3 MCP Server
Author: samuraikun

by samuraikun

Overview Schema Source Related Servers Discussions Admin

TypeScript

Remote

S3 MCP 서버

씨아이 트리비 스캔 npm 버전 npm 다운로드 라이센스: MIT Node.js 버전

S3 버킷 및 객체와 상호 작용하기 위한 도구를 제공하는 Amazon S3 모델 컨텍스트 프로토콜(MCP) 서버입니다.

https://github.com/user-attachments/assets/d05ff0f1-e2bf-43b9-8d0c-82605abfb666

개요

이 MCP 서버는 Claude와 같은 대규모 언어 모델(LLM)이 AWS S3 스토리지와 상호 작용할 수 있도록 지원합니다. 다음과 같은 도구를 제공합니다.

사용 가능한 S3 버킷 나열
버킷 내의 객체 나열
객체 내용 검색

이 서버는 TypeScript와 MCP SDK를 사용하여 구축되었으며, LLM이 S3와 상호 작용하는 안전하고 표준화된 방법을 제공합니다.

Related MCP server: AWS MCP Server

설치

필수 조건

Node.js 18 이상
npm 또는 yarn
AWS 자격 증명이 구성됨(환경 변수 또는 AWS 자격 증명 파일을 통해)
Docker(컨테이너화된 설정을 위한 선택 사항)

설정

npm을 통해 설치:

지엑스피1

소스에서 빌드하는 경우:

# Clone the repository
git clone https://github.com/samuraikun/aws-s3-mcp.git
cd aws-s3-mcp

# Install dependencies and build
npm install
npm run build

AWS 자격 증명 및 S3 액세스를 구성합니다.

AWS 구성을 사용하여 .env 파일을 만듭니다.

AWS_REGION=us-east-1
S3_BUCKETS=bucket1,bucket2,bucket3
S3_MAX_BUCKETS=5
AWS_ACCESS_KEY_ID=your-access-key
AWS_SECRET_ACCESS_KEY=your-secret-key

또는 이를 환경 변수로 설정합니다.

구성

다음 환경 변수를 사용하여 서버를 구성할 수 있습니다.

변하기 쉬운	설명	기본
`AWS_REGION`	S3 버킷이 있는 AWS 지역	`us-east-1`
`S3_BUCKETS`	허용된 S3 버킷 이름의 쉼표로 구분된 목록	(비어 있는)
`S3_MAX_BUCKETS`	목록에 반환할 버킷의 최대 개수	`5`
`AWS_ACCESS_KEY_ID`	AWS 액세스 키(기본 자격 증명을 사용하지 않는 경우)	(AWS 구성에서)
`AWS_SECRET_ACCESS_KEY`	AWS 비밀 키(기본 자격 증명을 사용하지 않는 경우)	(AWS 구성에서)

서버 실행

직접 Node.js 실행

Node.js를 사용하여 서버를 직접 실행할 수 있습니다.

# Using npx (without installing)
npx aws-s3-mcp

# If installed globally
npm install -g aws-s3-mcp
aws-s3-mcp

# If running from cloned repository
npm start

# Or directly
node dist/index.js

Docker 설정 🐳

Docker CLI 또는 Docker Compose를 사용하여 S3 MCP 서버를 Docker 컨테이너로 실행할 수 있습니다.

Docker CLI 사용

Docker 이미지를 빌드합니다.

docker build -t aws-s3-mcp .

환경 변수를 사용하여 컨테이너를 실행합니다.

# Option 1: Pass environment variables directly
docker run -d \
  -e AWS_REGION=us-east-1 \
  -e S3_BUCKETS=bucket1,bucket2 \
  -e S3_MAX_BUCKETS=5 \
  -e AWS_ACCESS_KEY_ID=your-access-key \
  -e AWS_SECRET_ACCESS_KEY=your-secret-key \
  --name aws-s3-mcp-server \
  aws-s3-mcp

# Option 2: Use environment variables from .env file
docker run -d \
  --env-file .env \
  --name aws-s3-mcp-server \
  aws-s3-mcp

컨테이너 로그를 확인하세요:

docker logs aws-s3-mcp-server

멈추고 용기를 제거하세요:

docker stop aws-s3-mcp-server
docker rm aws-s3-mcp-server

컨테이너는 Docker exec를 통해 Claude Desktop과 함께 사용하도록 설계되었기 때문에 어떤 포트도 노출하지 않으며, 직접 HTTP 연결을 사용하도록 설계되었다는 점에 유의하세요.

Docker Compose 사용

Docker 컨테이너를 빌드하고 시작합니다.

# Build and start the container
docker compose up -d s3-mcp

# View logs
docker compose logs -f s3-mcp

컨테이너를 멈추려면:

docker compose down

테스트를 위해 MinIO와 함께 Docker 사용

Docker Compose 설정에는 로컬 테스트를 위한 MinIO 서비스가 포함되어 있습니다.

# Start MinIO and the MCP server
docker compose up -d

# Access MinIO console at http://localhost:9001
# Default credentials: minioadmin/minioadmin

MinIO 서비스는 자동으로 두 개의 테스트 버킷( test-bucket-1 및 test-bucket-2 )을 생성하고 테스트를 위해 샘플 파일을 업로드합니다.

MCP Inspector에서 디버깅

MCP Inspector를 사용하여 서버를 디버깅하려면:

# Run inspector with local Node.js
sh run-inspector.sh

# Run inspector with Docker Compose and MinIO
sh run-inspector.sh --docker-compose

# Run inspector with Docker CLI (without Docker Compose)
sh run-inspector.sh --docker

--docker-compose 옵션을 사용하면 스크립트는 다음을 수행합니다.

MinIO 및 S3 MCP 서버 컨테이너가 실행 중이 아니면 시작합니다.
S3 MCP 서버에 연결된 MCP Inspector를 실행합니다.
그런 다음 로컬 MinIO 인스턴스에 대해 S3 도구를 테스트할 수 있습니다.

Claude Desktop에 연결

Claude Desktop과 함께 이 서버를 사용하려면:

Claude Desktop 구성 파일을 편집하세요.
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%\Claude\claude_desktop_config.json
구성에 S3 MCP 서버를 추가합니다.

{
  "mcpServers": {
    "s3": {
      "command": "npx",
      "args": ["aws-s3-mcp"],
      "env": {
        "AWS_REGION": "us-east-1",
        "S3_BUCKETS": "bucket1,bucket2,bucket3",
        "S3_MAX_BUCKETS": "5",
        "AWS_ACCESS_KEY_ID": "your-access-key",
        "AWS_SECRET_ACCESS_KEY": "your-secret-key"
      }
    }
  }
}

Claude Desktop용 Docker 옵션 🐳

MCP 서버에 실행 중인 Docker 컨테이너를 사용하도록 Claude Desktop을 구성할 수도 있습니다.

{
  "mcpServers": {
    "s3": {
      "command": "docker",
      "args": ["exec", "-i", "aws-s3-mcp-server", "node", "dist/index.js"],
      "env": {}
    }
  }
}

⚠️ 중요 전제 조건 : 이 Docker 구성이 작동하려면 Claude Desktop을 시작하기 전에 먼저 Docker 컨테이너를 빌드하고 실행 해야 합니다 .
# 1. First, build the Docker image (only needed once or after changes)
docker build -t aws-s3-mcp .

# 2. Then start the container (required each time before using with Claude)
# Using Docker Compose (recommended)
docker compose up -d s3-mcp

# Or using Docker CLI
docker run -d --name aws-s3-mcp-server --env-file .env aws-s3-mcp
실행 중인 컨테이너가 없으면 Claude Desktop에서 S3 도구를 사용하려고 할 때 오류가 표시됩니다.
위의 Docker 구성은 exec 사용하여 MCP 요청을 실행 중인 컨테이너로 직접 전송합니다. Claude가 네트워크 포트를 통하지 않고 컨테이너와 직접 통신하므로 포트 매핑이 필요하지 않습니다.

참고 : 구성( aws-s3-mcp-server )의 컨테이너 이름이 실행 중인 컨테이너의 이름과 일치하는지 확인하세요.

중요 : 위의 구성을 사용할 때 다음 사항에 유의하세요.
AWS_ACCESS_KEY_ID 및 AWS_SECRET_ACCESS_KEY 실제 자격 증명으로 바꾸세요.
S3_BUCKETS 에는 액세스를 허용하려는 버킷의 쉼표로 구분된 목록이 포함되어야 합니다.
AWS_REGION 버킷이 있는 지역으로 설정해야 합니다.

💣 Claude Desktop에서 오류가 발생하는 경우

Claude Desktop에서 위 구성과 관련하여 오류가 발생하는 경우 다음과 같이 절대 경로를 사용해 보세요.

# Get the path of node and aws-s3-mcp
which node
which aws-s3-mcp

{
  "globalShortcut": "",
  "mcpServers": {
    "s3": {
      "command": "your-absolute-path-to-node",
      "args": ["your-absolute-path-to-aws-s3-mcp/dist/index.js"],
      "env": {
        "AWS_REGION": "your-aws-region",
        "S3_BUCKETS": "your-s3-buckets",
        "S3_MAX_BUCKETS": "your-max-buckets",
        "AWS_ACCESS_KEY_ID": "your-access-key",
        "AWS_SECRET_ACCESS_KEY": "your-secret-key"
      }
    }
  }
}

사용 가능한 도구

리스트 버킷

서버가 액세스할 수 있는 사용 가능한 S3 버킷을 나열합니다. 이 도구는 표시되는 버킷을 제한하는 S3_BUCKETS 구성을 따릅니다.

매개변수: 없음

출력 예:

[
  {
    "Name": "my-images-bucket",
    "CreationDate": "2022-03-15T10:30:00.000Z"
  },
  {
    "Name": "my-documents-bucket",
    "CreationDate": "2023-05-20T14:45:00.000Z"
  }
]

목록 객체

지정된 S3 버킷에 있는 객체를 나열합니다.

매개변수:

bucket (필수): 객체를 나열할 S3 버킷의 이름
prefix (선택 사항): 객체를 필터링하기 위한 접두사(예: 폴더 경로)
maxKeys (선택 사항): 반환할 최대 개체 수

출력 예:

[
  {
    "Key": "sample.pdf",
    "LastModified": "2023-10-10T08:12:15.000Z",
    "Size": 2048576,
    "StorageClass": "STANDARD"
  },
  {
    "Key": "sample.md",
    "LastModified": "2023-10-12T15:30:45.000Z",
    "Size": 1536000,
    "StorageClass": "STANDARD"
  }
]

get-object

지정된 S3 버킷에서 객체를 검색합니다. 텍스트 파일은 일반 텍스트로 반환되고, 이진 파일은 제한된 세부 정보로 반환됩니다.

매개변수:

bucket (필수): S3 버킷의 이름
key (필수): 검색할 객체의 키(경로)

텍스트 출력 예시:

This is the content of a text file stored in S3.
It could be JSON, TXT, CSV or other text-based formats.

이진 출력 예:

Binary content (image/jpeg): base64 data is /9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRof...

보안 고려 사항

서버는 S3_BUCKETS 환경 변수에 지정된 버킷에만 액세스합니다.
AWS 자격 증명에는 버킷에 대한 적절한 권한이 있어야 합니다.
AWS 권한을 구성할 때 최소 권한 원칙을 사용하세요.
프로덕션 용도로는 특정 S3 권한이 있는 IAM 역할을 사용하는 것이 좋습니다.

Claude와 함께 사용

데스크톱 앱에서 Claude와 상호 작용할 때 다음과 같은 S3 작업을 수행하도록 요청할 수 있습니다.

"내 S3 버킷 모두 나열"
"내 문서 버킷에 있는 PDF 파일 요약"
"my-documents-bucket에서 README.txt 파일을 가져옵니다"

클로드는 적절한 MCP 도구를 사용하여 요청을 수행하고 결과를 보여드립니다.

특허

MIT

Available Tools

3 tools

get-objectB

Retrieve an object from an S3 bucket

ParametersJSON Schema

Name	Required	Description	Default
`bucket`	Yes	Name of the S3 bucket
`key`	Yes	Key (path) of the object to retrieve

TDQS

B3.3/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided; description does not disclose behavioral traits such as size limits, access requirements, error handling, or return format.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Description is concise with one sentence, but lacks depth; efficient but could include more context.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

No output schema and no description of return value (object content) or error behavior; incomplete for a retrieval tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% with clear descriptions for both parameters; description adds no additional meaning beyond schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Description uses specific verb 'Retrieve' and resource 'object from an S3 bucket', clearly distinguishing from sibling tools list-buckets and list-objects.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No explicit guidance on when to use or alternatives; usage is implied but not elaborated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

list-bucketsA

List available S3 buckets

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

A4/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description must disclose behavioral traits, but it only states 'List available S3 buckets' without mentioning any side effects, authorization, format, or limits. This is minimal transparency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A single, front-loaded sentence with no wasted words. Every part is essential.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's simplicity (no parameters, no output schema), the description adequately explains its purpose. It could mention the return format, but not strictly necessary.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The tool has zero parameters, so the description does not need to add param info. The baseline is 4, and the description's mention of 'available' adds slight context.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb 'list' and resource 'S3 buckets', clearly distinguishing it from sibling tools like get-object and list-objects.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

While no explicit when/when-not guidance is given, the context is clear: use this when you need to list available S3 buckets. Alternatives are not needed since siblings cover different resources.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

list-objectsB

List objects in an S3 bucket

ParametersJSON Schema

Name	Required	Description
`bucket`	Yes	Name of the S3 bucket
`prefix`	No	Prefix to filter objects (like a folder)
`maxKeys`	No	Maximum number of objects to return

TDQS

B3.3/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries full burden of behavioral disclosure. It does not mention pagination, performance implications, error handling, or that it is a read-only operation. Minimal behavioral insight beyond the action itself.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single sentence, very concise. However, it could include more useful information without becoming overly verbose.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given no output schema and no annotations, the description is incomplete. It does not describe the output format, pagination behavior, or error cases, leaving the agent with insufficient context.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% (all parameters have descriptions). The description adds no additional meaning beyond the schema; baseline score of 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description 'List objects in an S3 bucket' uses a specific verb and resource, clearly distinguishing it from siblings 'get-object' (gets a single object) and 'list-buckets' (lists all buckets).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No explicit guidance on when to use this tool vs alternatives, or how to apply parameters like prefix or maxKeys. The purpose is clear but the description lacks usage context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

3 tool updatesv1.0.0Jan 2, 2026
- Changedget-object1 field changed
  - removedInput schema / additionalProperties
```
Removed value: -false
```
- Changedlist-buckets1 field changed
  - removedInput schema / additionalProperties
```
Removed value: -false
```
- Changedlist-objects5 fields changed
  - removedInput schema / additionalProperties
```
Removed value: -false
```
  - addedInput schema / properties / maxKeys / anyOf
```
Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]
```
  - removedInput schema / properties / maxKeys / type
```
Removed value: -[
-  "number",
-  "null"
-]
```
  - addedInput schema / properties / prefix / anyOf
```
Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "type": "null"
+  }
+]
```
  - removedInput schema / properties / prefix / type
```
Removed value: -[
-  "string",
-  "null"
-]
```
3 tool updatesAug 18, 2025
- First observedget-object
- First observedlist-buckets
- First observedlist-objects

Deploy Server

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.7/5.0

Scored 2026-05-02 02:56 across 3 tools

Disambiguation5/5

Each tool targets a distinct S3 operation: retrieving objects, listing buckets, and listing objects. There is no overlap or ambiguity in their purposes.

Naming Consistency5/5

All tool names follow a consistent verb_noun pattern using snake_case (e.g., get-object, list-buckets), making them predictable and easy to understand.

Tool Count5/5

With 3 tools, the server is tightly scoped to basic read operations, avoiding unnecessary bloat while covering essential S3 actions for a minimal MCP server.

Completeness3/5

The server covers listing and retrieval but lacks write operations like put-object or delete-object, which are notable gaps for a general S3 server. Agents cannot perform updates or deletions.

Maintenance

ActivityInactive

ResponsivenessUnresponsive

Resources

Need Help?

Reddit Discussion

Related Servers

Related MCP Connectors

hithereiamaliff-mcp-nextcloudOAuth
ai.smithery
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Cloud Storage File Systems Workplace & Productivity
238
AWS Knowledge A
aws.api.global.knowledge-mcp
The AWS Knowledge MCP server is a fully managed remote Model Context Protocol server that provides real-time access to official AWS content in an LLM-compatible format. It offers structured access to AWS documentation, code samples, blog posts, What's New announcements, Well-Architected best practices, and regional availability information for AWS APIs and CloudFormation resources. Key capabilities include searching and reading documentation in markdown format, getting content recommendations, listing AWS regions, and checking regional availability for services and features.
Documentation Access Cloud Platforms Search
52
Agent Central A
to.agentcentral
Hosted Amazon Seller Central and Amazon Ads MCP server for Claude, ChatGPT, Cursor, and agents.
Customer Support E-commerce & Retail
9726
Sentry AOAuth
dev.sentry.mcp
Enable secure connectivity between Sentry issues and debugging data, and LLM clients, using a Model Context Protocol (MCP) server.
Monitoring Developer Tools Observability
3

Related MCP Servers

s3-tools
Cloud Storage Cloud Platforms
sofianhamiti
F
license
B
quality
D
maintenance
An MCP server that provides tools for interacting with AWS S3 buckets, enabling direct access to S3 operations through the Model Context Protocol.
Updated 2 years ago (2025-02-22 16:02 UTC)
1
-
AWS MCP Server
Cloud Platforms Cloud Storage Databases
rishikavikondala
A
license
A
quality
D
maintenance
A Model Context Protocol server implementation that enables Claude to perform AWS operations on S3 and DynamoDB services through natural language commands.
Updated 2 years ago (2025-04-08 01:43 UTC)
23
127
MIT
AWS MCP Server
Cloud Platforms Agent Orchestration Command Line
ihatesea69
F
license
C
quality
D
maintenance
A Model Context Protocol server allowing Claude AI to interact with AWS resources through natural language, enabling users to query and manage AWS services without using the traditional AWS Console or CLI.
Updated 10 months ago (2025-12-19 07:56 UTC)
3
6
-
s3-mcp
KonMam
A
license
A
quality
D
maintenance
A Model Context Protocol (MCP) server for AWS S3 integration using FastMCP and boto3. This server provides access to S3 functionality through MCP-compatible tools.
Updated a year ago (2025-07-20 07:32 UTC)
10
1
MIT

S3 MCP 서버

개요

설치

필수 조건

설정

구성

서버 실행

직접 Node.js 실행

Docker 설정 🐳

Docker CLI 사용

Docker Compose 사용

테스트를 위해 MinIO와 함께 Docker 사용

MCP Inspector에서 디버깅

Claude Desktop에 연결

Claude Desktop용 Docker 옵션 🐳

💣 Claude Desktop에서 오류가 발생하는 경우

사용 가능한 도구

리스트 버킷

목록 객체

get-object

보안 고려 사항

Claude와 함께 사용

특허

Available Tools

Tool Schema Changelog

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

Appeared in Searches