Package Version Check MCP

package-version-check-mcp

A MCP server that returns the latest stable versions of packages you use as dependencies in a variety of ecosystems, such as Python, NPM, Go, or GitHub Actions.

It also supports looking up the latest versions of almost 1000 tools, such as development runtimes like python, node, dotnet, development tools like gradle, and various DevOps tools like kubectl or terraform, via the mise-en-place tool.

Why do I need this?

Whenever an AI coding agents generates files that pin dependency versions, they insert outdated versions because their training happened weeks or months ago, and new dependency versions have been released since then. As a developer, it is annoying having to manually fix these outdated versions.

This MCP fixes this problem. Use it together with an MCP such as Context7 to avoid that your AI agent produces outdated code.

Features

Supported ecosystems / tools:

• Developer ecosystems:
• NPM - Node.js packages from registry.npmjs.org
• PyPI - Python packages from PyPI
• NuGet - .NET packages from NuGet
• Maven / Gradle - Java/Kotlin/Scala packages from Maven repositories (Maven Central, Google Maven, etc.)
• Go - Go modules from proxy.golang.org
• PHP - PHP packages from Packagist (used by Composer)
• Ruby - Ruby gems from rubygems.org
• Rust - Rust crates from crates.io
• Swift - Swift packages from GitHub repositories
• Dart - Dart packages from pub.dev
• DevOps ecosystems:
• Docker - Docker container images from Docker registries
• Helm - Helm charts from ChartMuseum repositories and OCI registries
• GitHub Actions - Actions hosted on GitHub.com, returning their current version, their inputs and outputs, and (optionally) their entire README with usage examples
• Terraform _Providers_ and _Modules_ - Providers & Modules from Terraform Registry, OpenTofu Registry, or custom registries
• Various _tools_ such as kubectl, terraform, gradle, maven, etc. supported by mise-en-place

Usage

Adding the MCP to Your Agent

There are three ways to make this MCP available to your AI coding agent:

Option 1: Use the Hosted Service (Easiest)

Point your agent to the free hosted service:

https://package-version-check-mcp.onrender.com/mcp

in (streamable) HTTP mode.

This is the quickest way to get started. Note that the hosted service may have rate limits from the underlying package registries.

Option 2: Run with uvx (for local use)

Use uvx to run the MCP server locally:

uvx package-version-check-mcp --mode=stdio

This automatically installs and runs the latest version from PyPI.

Requirements:

• You need the mise binary on PATH if you want to call the tools get_supported_tools or get_latest_tool_versions

Optional but recommended: Set the GITHUB_PAT environment variable to a GitHub Personal Access Token (no scopes required) to avoid GitHub API rate limits.

Option 3: Run with Docker (for local use)

Use the pre-built Docker image:

docker run --pull=always --rm -i ghcr.io/mshekow/package-version-check-mcp:latest --mode=stdio

Optional but recommended: Pass the GITHUB_PAT environment variable using -e GITHUB_PAT=your_token_here to avoid GitHub API rate limits.

Caching Configuration

To improve performance and reduce API calls to package registries, you can enable caching:

• PACKAGE_VERSION_CACHE_ENABLED: Set to true to enable caching (disabled by default)
• PACKAGE_VERSION_CACHE_TTL_SECONDS: Cache duration in seconds (default: 3600 / 1 hour)
• PACKAGE_VERSION_CACHE_MAX_SIZE_MB: Maximum cache size in MB (default: 64)

The cache is an in-memory TTL (Time-To-Live) cache. It resets when the MCP server restarts.

Configuring Your Agent

Once you've added the MCP server, you need to:

1. Enable the MCP tools in your agent's configuration. The available tools are documented below

1. Nudge the agent to use the MCP in your prompts. Most LLMs don't automatically invoke this MCP's tools without explicit guidance. Include instructions like:

• "Use MCP to get latest versions"
• "Check the latest package versions using the MCP tools"
• "Use get_latest_package_versions to find the current version"

In case you forgot to add this prompt and your agent generated code with _outdated_ versions, you can just ask your agent to update the versions afterwards (e.g., "Update the dependencies you just added to the latest version via MCP").

Available Tools

get_latest_package_versions

Fetches the latest versions of packages from various ecosystems.

Input:

• packages: Array of package specifications, where each item contains:
• ecosystem (required): Either "npm", "pypi", "nuget", "maven_gradle", "go", "php", "rubygems", "rust", "swift", "dart", "docker", "helm", "terraform_provider", or "terraform_module"
• package_name (required): The name of the package
• For npm: package name (e.g., "express")
• For pypi: package name (e.g., "requests")
• For nuget: package name (e.g., "Newtonsoft.Json")
• For maven_gradle: "[registry:] : " format (e.g., "org.springframework:spring-core"). If registry is omitted, Maven Central is assumed.
• For go: Absolute module identifier (e.g., "github.com/gin-gonic/gin")
• For php: Package name in "vendor/package" format (e.g., "monolog/monolog", "laravel/framework")
• For rubygems: Gem name (e.g., "rails", "devise")
• For rust: Crate name (e.g., "serde", "tokio")
• For swift: GitHub URL (e.g., "https://github.com/Alamofire/Alamofire.git" or "github.com/owner/repo.git"). Only github.com is supported.
• For dart: Package name from pub.dev (e.g., "http", "flutter")
• For docker: fully qualified image name including registry and namespace (e.g., "index.docker.io/library/busybox")
• For helm: Either ChartMuseum URL ("https://host/path/chart-name") or OCI reference ("oci://host/path/chart-name")
• For terraform_provider: "[registry/] / " format (e.g., "hashicorp/aws" or "registry.terraform.io/hashicorp/aws"). If registry is omitted, registry.terraform.io is assumed. Supports alternative registries like registry.opentofu.org.
• For terraform_module: "[registry/] / / " format (e.g., "terraform-aws-modules/vpc/aws" or "registry.terraform.io/terraform-aws-modules/vpc/aws"). If registry is omitted, registry.terraform.io is assumed. Supports alternative registries like registry.opentofu.org.
• version_hint (optional):
• For docker: tag compatibility hint (e.g., "1.36-alpine") to find the latest tag matching the same suffix pattern. If omitted, returns the latest semantic version tag.
• For helm (OCI only): tag compatibility hint similar to Docker
• For npm/pypi/nuget/maven_gradle/go/php/rubygems/rust/swift/dart/helm (ChartMuseum)/terraform_provider/terraform_module: not currently used

Output:

• result: Array of successful lookups with:
• ecosystem: The package ecosystem (as provided)
• package_name: The package name (as provided)
• latest_version: The latest version number (e.g., "1.2.4") or Docker tag
• digest: (optional) Package digest/hash if available. For Docker, this is the manifest digest (sha256).
• published_on: (optional) Publication date if available (not available for Docker)
• lookup_errors: Array of errors with:
• ecosystem: The package ecosystem (as provided)
• package_name: The package name (as provided)
• error: Description of the error

get_github_action_versions_and_args

Fetches the latest versions and metadata for GitHub Actions hosted on github.com.

Input:

• action_names (required): Array of action names in "owner/repo" format (e.g., ["actions/checkout", "docker/login-action"])
• include_readme (optional): Boolean (default: false), whether to include the action's README.md with usage instructions

Output:

• result: Array of successful lookups with:
• name: The action name (as provided)
• latest_version: The most recent Git tag (e.g., "v3.2.4")
• metadata: The action.yml metadata as an object with fields:
• inputs: Action input parameters
• outputs: Action outputs
• runs: Execution configuration
• readme: (optional) The action's README content if include_readme was true
• lookup_errors: Array of errors with:
• name: The action name (as provided)
• error: Description of the error

get_supported_tools

Returns a list of all tool names supported by the get_latest_tool_versions MCP tool.

This tool queries the mise registry to retrieve all available tool names that can be managed by mise.

Input:

• No parameters required

Output:

• Array of tool short names (e.g., ["1password", "act", "node", "python", ...])

get_latest_tool_versions

Fetches the latest stable versions of development and DevOps tools supported by mise-en-place.

This tool is for tools that are NOT part of language ecosystems like PyPI or NPM. For language ecosystem packages (including Terraform providers and modules), use get_latest_package_versions instead.

Use cases:

• gradle or maven: Pin the Gradle or Maven version in the distributionUrl in gradle-wrapper.properties or maven-wrapper.properties
• Example: distributionUrl=https://services.gradle.org/distributions/gradle-8.5-bin.zip
• Example: distributionUrl=https://repo.maven.apache.org/.../apache-maven-3.9.6-bin.zip
• terraform: Pin terraform.required_version in a file like version.tf or versions.tf
• Example: terraform { required_version = "~> 1.6.0" }
• kubectl or azure: Pin the version in a download URL called with curl or wget, e.g., in a Dockerfile
• Example: RUN curl -LO https://dl.k8s.io/release/v1.28.0/bin/linux/amd64/kubectl

To see all available tools, use the get_supported_tools tool.

Input:

• tool_names (required): Array of tool names (e.g., ["terraform", "gradle", "kubectl"])

Output:

• result: Array of successful lookups with:
• tool_name: The tool name (as provided)
• latest_version: The latest stable version number (e.g., "1.6.5")
• lookup_errors: Array of errors with:
• tool_name: The tool name (as provided)
• error: Description of the error

Why build yet another MCP

This MCP is certainly not the first one to tackle the "outdated dependency" problem. However, we feel that it has various advantages over other MCPs:

• We offer (far) better ecosystem coverage than other MCPs
• There is full test coverage, with automated dependency updates (powered by Renovate) and regular, automated release builds. In contrast, other projects are often vibe coded, have poor (or no) tests, and are already abandoned, because the authors were just messing around
• This MCP provides _several_ alternatives for how to run it locally (uvx or docker), or you can just use the free hosted offering (which other MCPs do not have)
• This MCP uses a minimal Docker/OCI image, hardened for security. SBOMs you generate with tools like Trivy are known to be correct, and the image is signed with Cosign (which allows you to verify its authenticity in case you want to self-host the MCP)

Development

Prerequisites

mise-en-place

The MCP server depends on the mise-en-place package for looking up tool versions. See https://mise.jdx.dev/installing-mise.html for installation instructions.

Running the Server Manually (For Development)

If you're developing or testing the MCP server locally, you can run it directly.

First, follow the Package management with Poetry -> Setup instructions to configure your virtual environments.

Next:

.poetry/bin/poetry run python -m package_version_check_mcp.main

Or if you have the .venv activated:

`