MCP Security Audit Server
# Security Audit Tool
[](https://smithery.ai/server/@qianniuspace/mcp-security-audit)
[](https://www.npmjs.com/package/mcp-security-audit)
[](https://opensource.org/licenses/MIT)
<a href="https://glama.ai/mcp/servers/jjnmdxzmeu">
<img width="380" height="200" src="https://glama.ai/mcp/servers/jjnmdxzmeu/badge" />
</a>
A powerful MCP (Model Context Protocol) Server that audits npm package dependencies for security vulnerabilities. Built with remote npm registry integration for real-time security checks.
## Features
- 🔍 Real-time security vulnerability scanning
- 🚀 Remote npm registry integration
- 📊 Detailed vulnerability reports with severity levels
- 🛡️ Support for multiple severity levels (critical, high, moderate, low)
- 📦 Compatible with npm/pnpm/yarn package managers
- 🔄 Automatic fix recommendations
- 📋 CVSS scoring and CVE references
### Installing via Smithery
To install Security Audit Tool for Claude Desktop automatically via [Smithery](https://smithery.ai/server/@qianniuspace/mcp-security-audit):
```bash
npx -y @smithery/cli install @qianniuspace/mcp-security-audit --client claude
```
### MCP Integration
#### Option 1: Using NPX (Recommended)
1. Add MCP configuration to Cline /Cursor:
```json
{
"mcpServers": {
"mcp-security-audit": {
"command": "npx",
"args": ["-y", "mcp-security-audit"]
}
}
}
```
#### Option 2: Download Source Code and Configure Manually
1. Clone the repository:
```bash
git clone https://github.com/qianniuspace/mcp-security-audit.git
cd mcp-security-audit
```
2. Install dependencies and build:
```bash
npm install
npm run build
```
3. Add MCP configuration to Cline /Cursor :
```json
{
"mcpServers": {
"mcp-security-audit": {
"command": "npx",
"args": ["-y", "/path/to/mcp-security-audit/build/index.js"]
}
}
}
```
## Configuration Screenshots
### Cursor Configuration

### Cline Configuration

## API Response Format
The tool provides detailed vulnerability information including severity levels, fix recommendations, CVSS scores, and CVE references.
### Response Examples
#### 1. When Vulnerabilities Found (Severity-response.json)
```json
{
"content": [{
"vulnerability": {
"packageName": "lodash",
"version": "4.17.15",
"severity": "high",
"description": "Prototype Pollution in lodash",
"cve": "CVE-2020-8203",
"githubAdvisoryId": "GHSA-p6mc-m468-83gw",
"recommendation": "Upgrade to version 4.17.19 or later",
"fixAvailable": true,
"fixedVersion": "4.17.19",
"cvss": {
"score": 7.4,
"vector": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:N"
},
"cwe": ["CWE-1321"],
"url": "https://github.com/advisories/GHSA-p6mc-m468-83gw"
},
"metadata": {
"timestamp": "2024-04-23T10:00:00.000Z",
"packageManager": "npm"
}
}]
}
```
#### 2. When No Vulnerabilities Found (no-Severity-response.json)
```json
{
"content": [{
"vulnerability": null,
"metadata": {
"timestamp": "2024-04-23T10:00:00.000Z",
"packageManager": "npm",
"message": "No known vulnerabilities found"
}
}]
}
```
## Development
For development reference, check the example response files in the `public` directory:
- [Severity-response.json](public/Severity-response.json) : Example response when vulnerabilities are found (transformed from npm audit API response)
- [no-Severity-response.json](public/no-Severity-response.json) : Example response when no vulnerabilities are found (transformed from npm audit API response)
Note: The example responses shown above are transformed from the raw npm audit API responses to provide a more structured format. The original npm audit API responses contain additional metadata and may have a different structure.
## Contributing
Contributions are welcome! Please read our [Contributing Guide](CODE_OF_CONDUCT.md) for details on our code of conduct and the process for submitting pull requests.
## License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## Author
ESX (qianniuspace@gmail.com)
## Links
- [GitHub Repository](https://github.com/qianniuspace/mcp-security-audit)
- [Issue Tracker](https://github.com/qianniuspace/mcp-security-audit/issues)
- [Changelog](CHANGELOG.md)
```
TDQS
Scored across 1 tool
With only one tool, there is no possibility of confusion between tools. The tool's purpose is clearly stated in its name and description, making it unambiguous.
A single tool name follows a clear verb_noun pattern (audit_nodejs_dependencies) which is consistent with typical MCP naming conventions. There are no other names to compare, but the pattern is clear.
Having only one tool is extremely thin for a security audit server. Auditing dependencies likely requires supporting operations like resolving dependency trees, checking for updates, or generating reports, which are missing.
The server's purpose appears to be auditing Node.js dependencies, but it only offers a single operation that audits dependencies without coverage for other lifecycle steps like viewing audit results, fixing vulnerabilities, or scanning different scopes.