Serverssammcj/mcp-package-version

sammcj/mcp-package-version

Created on 2025-03-22

Open Source Visit Server

An MCP server that provides LLMs with the latest stable package versions when coding

Information

Package Version MCP Server

An MCP server that provides tools for checking latest stable package versions from multiple package registries:

npm (Node.js/JavaScript)
PyPI (Python)
Maven Central (Java)
Go Proxy (Go)
AWS Bedrock (AI Mdels)
Docker Hub (Container Images)
GitHub Container Registry (Container Images)

This server helps LLMs ensure they're recommending up-to-date package versions when writing code.

Screenshot

tooling with and without mcp-package-version

Running

Installing via Smithery

To install Package Version for Claude Desktop automatically via Smithery:

npx -y @smithery/cli install mcp-package-version --client claude

Configure MCP Settings

Add the following to your MCP settings file:

{
  "mcpServers": {
    "package-version": {
      "command": "npx",
      "args": ["-y", "mcp-package-version"]
    }
  }
}

If you are behind a corporate proxy which MITMs your traffic, you may need to additionally specify the proxy CA cert bundle:

{
  "mcpServers": {
    "package-version": {
      "command": "npx",
      "args": ["-y", "mcp-package-version"],
      "env": {
        "NODE_EXTRA_CA_CERTS": "/path/to/mitm/cert.pem"
      }
    }
  }
}

For the Cline VSCode Extension this will be ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
For Claude Desktop ~/Library/Application\ Support/Claude/claude_desktop_config.json
For GoMCP ~/.config/gomcp/config.yaml

Tools

1. JavaScript/Node.js

check_npm_versions

Check latest stable versions for npm packages from a package.json dependencies object.

use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_npm_versions",
  arguments: {
    dependencies: {
      "express": "^4.17.1",
      "react": "^17.0.2"
    }
  }
});

2. Python

check_python_versions

Check latest stable versions for Python packages from requirements.txt entries.

check_pyproject_versions

Check latest stable versions for Python packages from pyproject.toml.

use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_pyproject_versions",
  arguments: {
    dependencies: {
      dependencies: {
        "requests": "^2.28.0",
        "pandas": ">=1.5.0"
      },
      "optional-dependencies": {
        "test": {
          "pytest": ">=7.0.0"
        }
      },
      "dev-dependencies": {
        "black": "^22.0.0"
      }
    }
  }
});

use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_python_versions",
  arguments: {
    requirements: [
      "requests==2.26.0",
      "pandas>=1.3.0"
    ]
  }
});

3. Go

check_go_versions

Check latest stable versions for Go packages from go.mod.

use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_go_versions",
  arguments: {
    dependencies: {
      module: "example.com/mymodule",
      require: [
        {
          path: "github.com/gin-gonic/gin",
          version: "v1.7.0"
        }
      ],
      replace: [
        {
          old: "github.com/old/pkg",
          new: "github.com/new/pkg",
          version: "v2.0.0"
        }
      ]
    }
  }
});

4. Java

But seriously, don't write Java in 2025.

check_maven_versions

Check latest stable versions for Java packages from pom.xml.

use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_maven_versions",
  arguments: {
    dependencies: [
      {
        groupId: "org.springframework.boot",
        artifactId: "spring-boot-starter-web",
        version: "2.7.0",
        scope: "compile"
      }
    ]
  }
});

check_gradle_versions

Check latest stable versions for Java packages from build.gradle.

use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_gradle_versions",
  arguments: {
    dependencies: [
      {
        configuration: "implementation",
        group: "com.google.guava",
        name: "guava",
        version: "31.0-jre"
      }
    ]
  }
});

5. check_package_versions

Bulk check latest stable versions for multiple packages from npm and PyPI.

use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_package_versions",
  arguments: {
    packages: [
      { name: "react", registry: "npm" },
      { name: "requests", registry: "pypi" },
      { name: "typescript", registry: "npm", currentVersion: "5.0.0" }
    ]
  }
});

6. AWS Bedrock

check_bedrock_models

Search, list, and get information about Amazon Bedrock AI models.

// List all available Bedrock models
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_bedrock_models",
  arguments: {
    action: "list"
  }
});

// Search for specific models
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_bedrock_models",
  arguments: {
    action: "search",
    query: "claude",
    provider: "anthropic"
  }
});

// Get a specific model by ID
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_bedrock_models",
  arguments: {
    action: "get",
    modelId: "anthropic.claude-3-sonnet-20240229-v1:0"
  }
});

get_latest_bedrock_model

Get the latest Claude Sonnet model from Amazon Bedrock (best for coding tasks).

use_mcp_tool({
  server_name: "package-version",
  tool_name: "get_latest_bedrock_model",
  arguments: {}
});

7. Docker Container Images

check_docker_tags

Check available tags for Docker container images from Docker Hub, GitHub Container Registry, or custom registries.

// Check Docker Hub images
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_docker_tags",
  arguments: {
    image: "nginx",
    limit: 5
  }
});

// Check GitHub Container Registry images
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_docker_tags",
  arguments: {
    image: "ghcr.io/owner/repo",
    registry: "ghcr"
  }
});

// Check custom registry images
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_docker_tags",
  arguments: {
    image: "my-image",
    registry: "custom",
    customRegistry: "registry.example.com"
  }
});

// Filter tags using regex patterns
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_docker_tags",
  arguments: {
    image: "node",
    filterTags: ["^18", "^20"],
    includeDigest: true
  }
});

Guidelines for LLMs

When writing code that includes package dependencies, LLMs should:

Choose the Right Tool for the Job
- Use language-specific tools for detailed dependency management:
  - check_npm_versions for package.json
  - check_python_versions for requirements.txt
  - check_pyproject_versions for pyproject.toml
  - check_maven_versions for pom.xml
  - check_gradle_versions for build.gradle
  - check_go_versions for go.mod
- Use check_package_versions for quick bulk checks across npm and PyPI
- Use AWS Bedrock tools for AI model information:
  - check_bedrock_models to search, list, or get specific model information
  - get_latest_bedrock_model to get the latest Claude Sonnet model (best for coding tasks)
- Use Docker container image tools:
  - check_docker_tags to find available tags for Docker images from Docker Hub, GitHub Container Registry, or custom registries
Always Check Versions Before Writing
- Before writing a package.json or requirements.txt file, use the appropriate tool to check latest versions
- Use the bulk check tool when dealing with multiple packages
- Consider the project's needs when deciding whether to use exact versions or version ranges

Package.json Best Practices

// Before writing package.json, check versions
const versions = await use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_package_versions",
  arguments: {
    packages: [
      { name: "express", registry: "npm" },
      { name: "react", registry: "npm" }
    ]
  }
});

// Use the returned versions in package.json
{
  "dependencies": {
    "express": "^{express.latestVersion}",
    "react": "^{react.latestVersion}"
  }
}

Requirements.txt Best Practices

// Before writing requirements.txt, check versions
const versions = await use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_package_versions",
  arguments: {
    packages: [
      { name: "requests", registry: "pypi" },
      { name: "pandas", registry: "pypi" }
    ]
  }
});

// Use the returned versions in requirements.txt
requests=={requests.latestVersion}
pandas=={pandas.latestVersion}

Version Range Considerations
- For applications: Consider using exact versions (= for Python, no prefix for npm)
- For libraries: Consider using compatible ranges (>= for Python, ^ for npm)
- Always document version choices in comments
Error Handling
- If version check fails for a package, document it in comments
- Consider falling back to known stable versions if checks fail
- Warn users about any packages that couldn't be verified

Example Integrations

Here's how an LLM should approach creating new projects with different package managers:

Node.js Project

// 1. Check npm package versions
const versions = await use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_npm_versions",
  arguments: {
    dependencies: {
      "express": "^4.17.1",
      "typescript": "~4.5.0"
    }
  }
});

// 2. Use the versions in package.json
write_to_file({

  path: "package.json",
  content: {
    "name": "my-project",
    "version": "1.0.0",
    "dependencies": {
      "express": `^${versions.find(p => p.name === 'express').latestVersion}`,
      "typescript": `^${versions.find(p => p.name === 'typescript').latestVersion}`
    }
  }
});

Python Project with pyproject.toml

// 1. Check Python package versions
const versions = await use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_pyproject_versions",
  arguments: {
    dependencies: {
      dependencies: {
        "requests": "^2.28.0",
        "pandas": ">=1.5.0"
      },
      "dev-dependencies": {
        "pytest": ">=7.0.0"
      }
    }
  }
});

// 2. Use the versions in pyproject.toml
write_to_file({
  path: "pyproject.toml",
  content: `
[project]
name = "my-project"
version = "1.0.0"
dependencies = [
    "requests>=${versions.find(p => p.name === 'requests').latestVersion}",
    "pandas>=${versions.find(p => p.name === 'pandas').latestVersion}"
]

[project.optional-dependencies]
test = [
    "pytest>=${versions.find(p => p.name === 'pytest (dev)').latestVersion}"
]
`
});

Go Project

// 1. Check Go package versions
const versions = await use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_go_versions",
  arguments: {
    dependencies: {
      module: "example.com/mymodule",
      require: [
        {
          path: "github.com/gin-gonic/gin",
          version: "v1.7.0"
        }
      ]
    }
  }
});

// 2. Use the versions in go.mod
write_to_file({
  path: "go.mod",
  content: `
module example.com/mymodule

go 1.21

require (
    github.com/gin-gonic/gin ${versions.find(p => p.name === 'github.com/gin-gonic/gin').latestVersion}
)
`
});

Java Project with Maven

// 1. Check Maven package versions
const versions = await use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_maven_versions",
  arguments: {
    dependencies: [
      {
        groupId: "org.springframework.boot",
        artifactId: "spring-boot-starter-web",
        version: "2.7.0"
      }
    ]
  }
});

// 2. Use the versions in pom.xml
write_to_file({
  path: "pom.xml",
  content: `
<project>
    <modelVersion>4.0.0</modelVersion>
    <groupId>com.example</groupId>
    <artifactId>my-project</artifactId>
    <version>1.0.0</version>

    <dependencies>
        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-starter-web</artifactId>
            <version>${versions.find(p => p.name === 'org.springframework.boot:spring-boot-starter-web').latestVersion}</version>
        </dependency>
    </dependencies>
</project>
`
});

This ensures that new projects always start with the latest stable versions of packages.

Example System Prompt

When writing code that includes package dependencies, you must:

1. Before writing any dependency files, check latest stable versions using the package-version MCP server:

// For Node.js (package.json):
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_npm_versions",
  arguments: {
    dependencies: {
      "package-name": "version"
    }
  }
});

// For Python (requirements.txt):
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_python_versions",
  arguments: {
    requirements: [
      "package-name==version"
    }
  }
});

// For Python (pyproject.toml):
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_pyproject_versions",
  arguments: {
    dependencies: {
      dependencies: {
        "package-name": "version"
      }
    }
  }
});

// For Java Maven (pom.xml):
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_maven_versions",
  arguments: {
    dependencies: [
      {
        groupId: "group",
        artifactId: "artifact",
        version: "version"
      }
    ]
  }
});

// For Java Gradle (build.gradle):
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_gradle_versions",
  arguments: {
    dependencies: [
      {
        configuration: "implementation",
        group: "group",
        name: "name",
        version: "version"
      }
    ]
  }
});

// For Go (go.mod):
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_go_versions",
  arguments: {
    dependencies: {
      module: "module-name",
      require: [
        {
          path: "package-path",
          version: "version"
        }
      ]
    }
  }
});

// For AWS Bedrock models:
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_bedrock_models",
  arguments: {
    action: "search",
    query: "claude",
    provider: "anthropic"
  }
});

// For getting the latest Claude Sonnet model:
use_mcp_tool({
  server_name: "package-version",
  tool_name: "get_latest_bedrock_model",
  arguments: {}
});

// For Docker container images:
use_mcp_tool({
  server_name: "package-version",
  tool_name: "check_docker_tags",
  arguments: {
    image: "image-name",
    registry: "dockerhub", // or "ghcr" or "custom"
    customRegistry: "registry.example.com", // required when registry is "custom"
    limit: 10, // optional, default is 10
    filterTags: ["regex-pattern"], // optional
    includeDigest: false // optional, default is false
  }
});

2. Use the returned latest versions in your dependency files:
   - For applications: Use exact versions
   - For libraries: Use compatible ranges
     - npm: ^ for minor updates, ~ for patch updates
     - Python: >= for compatible versions, == for exact versions
     - Java: Use the version directly (Maven/Gradle handle ranges differently)
     - Go: Use semantic version prefixes (e.g., v1.2.3)
   - Document any version-specific requirements in comments

3. If version checks fail:
   - Document it in comments
   - Use known stable versions as fallback
   - Consider project requirements and compatibility

Example system prompt for users:

When writing code that includes dependencies, you must check latest stable versions using the package-version MCP server before writing any dependency files (package.json, requirements.txt, pyproject.toml, pom.xml, build.gradle, go.mod). Use exact versions for applications and appropriate version ranges for libraries based on the package manager's conventions. Document any version-specific requirements or failed checks in comments. For AI model information, use the AWS Bedrock tools to search, list, or get specific model details. For Docker container images, use the check_docker_tags tool to find available tags from Docker Hub, GitHub Container Registry, or custom registries.

Development

Clone and Install Dependencies

git clone https://github.com/sammcj/mcp-package-version.git
cd mcp-package-version
npm i

Build the Server
```
npm run build
```
Development Workflow
- Use npm run watch for development to automatically rebuild on changes
- Use npm run build for production builds

Release Process

# 1. Make your changes
vim src/your-file.ts

# 2. Commit your changes
git add .
git commit -m "feat: your new feature"

# 3. Run bump command (this will):
# - Update version in package.json
# - Update CHANGELOG.md
# - Commit changes
# - Push to GitHub
npm run bump

# GitHub Actions will then:
# - Create a git tag
# - Create a GitHub release
# - Publish to npm (when triggered manually)

Manual npm Publishing

# To trigger a manual npm publish
gh workflow run publish.yml

No environment variables are required as this server uses public registries and documentation sites:

npm registry (registry.npmjs.org)
PyPI (pypi.org)
Go Proxy (proxy.golang.org)
Maven Central (search.maven.org)
AWS Bedrock documentation (docs.aws.amazon.com/bedrock)
Docker Hub (hub.docker.com)
GitHub Container Registry (ghcr.io)

License

MIT

Recommended MCP Servers

toolhive

Run and manage MCP servers easily and securely

A MCP Server for APK Tool (Part of Android Reverse Engineering MCP Suites)

attestable-mcp-server

Verify that any MCP server is running the intended and untampered code via hardware attestation.

Plugin for JADX to integrate MCP server

onepassword-mcp-server

An MCP server that enables secure credential retrieval from 1Password to be used by Agentic AI

An MCP Server for pyATS (experimental)

MCP server for AWS Cloud Development Kit (CDK) best practices, infrastructure as code patterns, and security compliance with CDK Nag.

mcp-cost-analysis-mcp-server

MCP server for analyzing AWS service costs and generating cost reports.

chaitin-ip-intelligence-search-tool

chaitin ip intelligence search tool for mcp

BloodHound-MCP-AI is integration that connects BloodHound with AI through Model Context Protocol, allowing security professionals to analyze Active Directory attack paths using natural language instead of complex Cypher queries.