Socket MCP Server
Connect Socket to your AI agents with the Socket MCP server.
Updated
What is Socket MCP?
The Socket MCP server is a Model Context Protocol server that connects Socket to AI agents and assistants as a set of tools. With it connected, an agent can work with Socket (software supply chain security — scan dependencies for risks and malware) directly — instead of you copying data in and out by hand.
It is a remote MCP server that uses no authentication, so you point your client at the server URL, authorize once, and the Socket tools become available to the model.
Connect the Socket MCP server
Connect Socket MCP to Claude Code
- 1
Add the server
Use `claude mcp add` with the Socket MCP server URL (or its `npx` command), following the official docs.
- 2
Authorize
Complete authentication (no authentication) so Claude Code can call the server.
- 3
Verify
Ask Claude to use a Socket tool to confirm the connection is live.
Connect Socket MCP to Cursor
- 1
Open MCP settings
In Cursor, go to Settings → MCP → Add new server.
- 2
Add Socket
Provide the Socket MCP server URL or command and complete authentication.
- 3
Test
Reference Socket from chat to confirm Cursor can reach the tools.
Sample use cases
Using Socket MCP with Gamut
On Gamut, the Socket MCP server becomes a tool your agents use inside an automated, event-driven workflow — triggered by an event or a schedule, acting through Socket and your other connected MCP servers, with the connection managed by Gamut rather than a local config file.
Frequently asked questions
What package ecosystems does Socket MCP support?
Socket supports npm (JavaScript/TypeScript), PyPI (Python), and Go modules. Coverage is broadest for npm, with Python and Go expanding.
How is Socket different from npm audit or Snyk?
Socket focuses on supply chain attacks (malicious code, typosquatting, compromised maintainers) rather than just known CVEs. It analyzes package behavior, not just vulnerability databases.
Does Socket MCP scan transitive dependencies?
Yes. Socket analyzes your full dependency tree, including transitive dependencies, to catch risks that direct dependency audits miss.
Is a paid Socket plan required?
Socket offers a free tier for public repos and limited private repo scans. Paid plans add higher scan limits, organization policies, and priority support.
Can I use Socket MCP in CI/CD pipelines?
Yes. Many teams use Socket's agent integration to automatically scan dependencies on every PR or deploy, blocking risky packages before they reach production.
Does Socket detect malicious packages?
Yes. Socket's core capability is detecting malicious packages — it analyzes runtime behavior (network calls, filesystem access, data exfiltration) to catch threats that CVE databases miss.
Can I customize risk thresholds?
Yes. Socket's policy engine lets you set custom rules — acceptable risk scores, allowed/blocked packages, required maintainer counts — tailored to your organization's security posture.