What does the Semgrep plugin do?

It lets the agent scan its own code for security bugs while it writes, catching vulnerabilities in real time and guiding it to write secure code from the start.

Does it scan code automatically?

Yes, the plugin catches security vulnerabilities in real-time as the agent writes code.

semgrep

PackSecurity

Semgrep is a plugin that lets an AI agent check the code it writes for security vulnerabilities as it works. The semgrep claude plugin scans the agent's own output in real time and guides it toward writing secure code from the start, rather than fixing issues after the fact.

Unavailable. Delivery for this kind is on the roadmap — not serving yet.

Have Semgrep available in the working environment.

What your AI can do with it

Scans code the agent writes for security bugs in real time
Guides the agent to write secure code from the start
Lets the agent scan its own code while it writes

Getting started

Have Semgrep available in the working environment.
Add the Semgrep plugin to the agent's configuration.
Start a coding session; the agent will scan its own code for security bugs as it writes.

Signals

GitHub stars: 37k
Forks: 4k
Last commit: Sep 2026

Questions

What does the Semgrep plugin do?: It lets the agent scan its own code for security bugs while it writes, catching vulnerabilities in real time and guiding it to write secure code from the start.
Does it scan code automatically?: Yes, the plugin catches security vulnerabilities in real-time as the agent writes code.

Advanced

Item type: plugin
Key: anthropics-claude-plugins-official-semgrep
Source: github.com/anthropics/claude-plugins-official

github.com/anthropics/claude-plugins-official