- VULNERABILITY
- TREE
- REPORT
- HUMAN
Short answer
How can I scan my dependency manifests using AI?
Your AI triggers a scan of your project files like package.json to map out your entire dependency tree. It then pulls security reports to pinpoint specific vulnerabilities. You get a clear view of what's broken and what needs fixing.
Security outcomes
Where risks are identified
What your AI produces after a scan.
VULNERABILITY
Identified risks
The AI finds specific security holes in your dependencies. You get a list of exactly which packages are unsafe.
TREE
Dependency mapping
The AI traces how packages connect to each other. This shows you how a single bad package affects your whole project.
REPORT
Detailed findings
The AI pulls full reports from Socket.dev. These reports contain the technical details you need to triage issues.
HUMAN
Manual remediation
The AI identifies the problems, but you decide which versions to upgrade. You handle the actual code changes and PRs.
The workflow
What your AI does when the manifest arrives.
The AI handles the heavy lifting of scanning and data retrieval.
Initiate the scan
Your AI starts the process by looking at your manifest files to build a scan profile.
create_scanCheck scan status
The AI monitors the progress of the analysis to ensure the data is ready.
get_scanRetrieve security data
Once the scan finishes, the AI pulls the specific vulnerability details.
get_reportSummarize findings
The AI uses the scan data to explain the risks in plain English.
get_report
Try it
Copy these to start.
Use these prompts to kick off a security check.
Starting points
Swap the bracketed text for your actual file names or repository paths.
Socket.dev Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_R4rzC31YO5lAo6xekKbMZrPuutcnC2L5364MbwbS/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Socket.dev capabilities are ready to use.
{
"mcpServers": {
"socketdev-dependency-security-mcp": {
"url": "https://edge.vinkius.com/vk_preview_R4rzC31YO5lAo6xekKbMZrPuutcnC2L5364MbwbS/mcp"
}
}
}Scan the package.json in my current directory.
Run a dependency scan for the project in ./frontend/package.json.
Get the latest security report for my current project scan.
Check the dependency tree for the manifest in my root folder.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Socket.dev once, then ask.
Just link your account once. Your credentials stay encrypted, and you can immediately start asking your AI to scan files.
Connect Socket.dev to your AIFAQ
How the scanning behaves
- 01
Can the AI fix my package.json file automatically?
No. The AI identifies the vulnerabilities and suggests updates, but you must manually edit the files or run the update commands.
- 02
Does the AI scan my entire codebase?
No. It scans the dependency manifests like package.json to analyze the libraries you use, not your custom application logic.
- 03
Will the AI delete any of my dependencies?
No. The AI only reads and analyzes data. It has no capability to modify or delete files in your repository.
- 04
How does the AI know which vulnerabilities exist?
The AI uses the Socket.dev tools to pull real-time security data based on the manifest it scans.
- 05
Can I run multiple scans at once?
Yes, the AI can trigger new scans for different manifest files as you request them.
More questions about Socket.dev? See everything the Socket.dev Connector can do
Connect Socket.dev to Claude, Cursor, ChatGPT & more
