Positive Technologies has updated its PT Application Inspector plugins for JetBrains IDE and Visual Studio Code to improve security in AI-assisted development. The tools are free and now integrate with AI agents via the Model Context Protocol (MCP).
The updated plugins support various AI agents, including Claude Code and Codex. Agents can now independently trigger security scans for entire projects or specific code changes. This integration aims to catch vulnerabilities that AI models might introduce during the coding process.
The new AI Hooks feature allows agents to scan only the modified code before finalizing a task, eliminating the need for a full project scan. Additionally, the 'How to Fix' function, which provides actionable remediation suggestions, is now available in Visual Studio Code.
Developers can connect any AI model with a compatible API to the plugin. This includes models hosted on private servers, ensuring that source code remains secure and is not transmitted to external third-party services.





