pkg-booper
Provides AI assistants with tools to check npm package versions and security heuristics, including version lookups, signal-based analysis, and batch analysis of package.json files.
README
pkg-booper
Boop your dependencies' snoots.
An MCP server that gives AI assistants tools to check up-to-date npm package versions, as well as gleaning other security heuristics. It's not a super comprehensive analysis, just a lil' snoot boop.
What is this?
pkg-booper is a Model Context Protocol (MCP) server that provides npm package intelligence for AI-assisted development. It fills the gap where documentation tools exist but version/health checking doesn't.
Key features:
- Quick version lookups for any npm package
- Security signal analysis using transparent indicators
- Batch analysis of entire
package.jsonfiles - Caching for fast repeated lookups
Installation
Run directly with npx:
npx pkg-booper
Or install globally:
npm install -g pkg-booper
pkg-booper
MCP Client Configuration
Claude Desktop
Add to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"pkg-booper": {
"command": "npx",
"args": ["-y", "pkg-booper"]
}
}
}
Cursor
Add to your Cursor MCP settings (.cursor/mcp.json):
{
"mcpServers": {
"pkg-booper": {
"command": "npx",
"args": ["-y", "pkg-booper"]
}
}
}
Claude Code
Add to your Claude Code MCP settings file (.mcp.json):
{
"mcpServers": {
"pkg-booper": {
"command": "npx",
"args": ["-y", "pkg-booper"]
}
}
}
Cline
Add to your Cline MCP settings:
{
"mcpServers": {
"pkg-booper": {
"command": "npx",
"args": ["-y", "pkg-booper"]
}
}
}
Tools
pkg-booper-boop
Quick version lookup for an npm package.
Input:
{
"package": "lodash",
"skipCache": false
}
Output:
{
"success": true,
"package": "lodash",
"latest": "4.17.21",
"tags": {
"latest": "4.17.21"
},
"summary": "lodash@4.17.21"
}
pkg-booper-sniff
Signal-based security analysis of an npm package. Returns information that lets the AI reason about the combination of indicators.
Input:
{
"package": "express",
"version": "4.18.2",
"compareVersions": true,
"skipCache": false
}
Output:
{
"success": true,
"package": "express",
"version": "4.18.2",
"facts": {
"age": { "days": 4500, "created": "2010-12-29" },
"lastPublish": { "days": 180, "version": "4.18.2" },
"weeklyDownloads": 30000000,
"maintainerCount": 3,
"versionCount": 275,
"hasTypes": true,
"hasRepository": true
},
"definitiveIndicators": [],
"behavioralSignals": [],
"info": ["Popular package: 30M weekly downloads"],
"summary": "express@4.18.2: No concerning signals detected"
}
Signal Categories:
definitiveIndicators: Critical security issues (malware markers, deprecated status)behavioralSignals: Patterns that warrant investigation (new package, single maintainer, suspicious scripts)info: Neutral observations (popularity, age, features)
pkg-booper-big-sniff-file
Batch analyze all dependencies from a package.json file.
Input:
{
"path": "/path/to/package.json",
"includeDevDependencies": true,
"skipCache": false
}
Output:
{
"success": true,
"packageJsonPath": "/path/to/package.json",
"results": {
"lodash": { "success": true, "package": "lodash", "..." },
"express": { "success": true, "package": "express", "..." }
},
"summary": {
"total": 15,
"clean": 14,
"suspicious": 1,
"outdated": 3,
"definiteThreats": 0,
"failed": 0
},
"humanSummary": "Analyzed 15 packages: 14 clean, 1 suspicious, 0 threats"
}
pkg-booper-big-sniff-list
Batch analyze a provided list of dependencies.
Input:
{
"dependencies": {
"lodash": "^4.17.21",
"express": "^4.18.2",
"axios": "^1.6.0"
},
"skipCache": false
}
Output: Same format as pkg-booper-big-sniff-file.
pkg-booper-clear-cache
Clear cached npm package data.
Input:
{
"package": "lodash",
"type": "all"
}
package: Optional. Clear cache for specific package, or omit for all.type:"all"|"versions"|"health"|"popular"
Output:
{
"success": true,
"cleared": ["lodash:versions", "lodash:health"],
"summary": "Cleared 2 cache entries for lodash"
}
Configuration
pkg-booper can be configured via YAML files, environment variables, or programmatic overrides.
Configuration File Locations
Files are checked in this order (first found wins):
PKG_BOOPER_CONFIGenvironment variable path./pkg-booper.yaml(project directory)~/.config/pkg-booper/config.yaml(user config)
Example Configuration
# pkg-booper.yaml
cache:
directory: ~/.cache/pkg-booper
ttl: 86400 # 24 hours in seconds
popularPackagesTtl: 604800 # 7 days for popular packages
thresholds:
levenshtein: 2 # Typosquat detection sensitivity
lowDownloads: 100 # Weekly downloads below this trigger a signal
scoring:
weights:
maturity: 0.15
popularity: 0.25
maintenance: 0.25
trust: 0.25
completeness: 0.10
github:
token: ${GITHUB_TOKEN} # Interpolates from environment
timeout: 10000 # Request timeout in ms
rateLimit:
npm:
requestsPerSecond: 10
github:
requestsPerMinute: 30
debug: false
Environment Variables
| Variable | Description | Default |
|---|---|---|
GITHUB_TOKEN |
GitHub API token for enhanced rate limits | - |
PKG_BOOPER_CONFIG |
Path to custom config file | - |
PKG_BOOPER_CACHE_DIR |
Cache directory | ~/.cache/pkg-booper |
PKG_BOOPER_CACHE_TTL |
Cache TTL in seconds | 86400 |
PKG_BOOPER_TIMEOUT |
Request timeout in ms | 10000 |
PKG_BOOPER_DEBUG |
Enable debug logging | false |
Examples
AI Assistant Usage
User: "What's the latest version of React?"
AI uses pkg-booper-boop:
React is currently at version 18.2.0.
User: "Analyze the security of this package.json"
AI uses pkg-booper-big-sniff-file:
I analyzed 23 dependencies:
- 21 packages look clean
- 2 packages have behavioral signals worth noting:
- some-package: New package (< 30 days old), single maintainer
- another-pkg: Significantly outdated (2 major versions behind)
No definitive threats detected, but you may want to review the flagged packages.
User: "Is 'lod-ash' a legit package?"
AI uses pkg-booper-sniff:
Warning: 'lod-ash' appears to be a typosquat of 'lodash':
- Very similar name (Levenshtein distance: 1)
- Much lower download count than lodash
- Created recently compared to lodash
I'd recommend using the official 'lodash' package instead.
License
This is licensed under the AGPLv3.
Why? I feel like information deserves to be free. Code is a sort of information, and I know the AGPL is pretty good at encouraging sharing. That, and I think of the code I put together with the help of generative AI less like an original composition, and more like a silly mashup.
I don't feel comfortable publishing my mashups under anything less open ;)
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.