Recent Updates
Agentic AI Interface
The AI-Assistant now supports an agentic interaction model — instead of just answering single prompts, it can work through multi-step tasks autonomously. Ask it to create a full test from a description, generate questions, write explanations, and organize them by topic — all in one conversation without manual step-by-step instructions.Improved Question Extraction from PDF
Question extraction from PDF documents has been significantly improved. The AI now handles complex layouts, multi-column PDFs, and documents with mixed formatting much more reliably. Answer options and correct answer markers are recognized more accurately, reducing the amount of manual cleanup needed after import.Question Recognition from Screenshots
You can now paste or upload screenshots directly into the AI chat and it will recognize questions from the image. This makes it easy to import questions from slides, printed materials, textbook photos, or any other visual source — without needing to retype anything.Free Answer Questions with AI Review
A new question type: free answer. Applicants write a text response instead of picking from options. You define AI review criteria when creating the question, and after the assessment is submitted, AI automatically evaluates the response and determines if it’s correct. This is useful for assessments where you want to test understanding in the applicant’s own words — process descriptions, explanations, situational responses — without reviewing every answer manually.MCP Server Integration
TestsPlatform now supports the Model Context Protocol (MCP), allowing AI tools like ChatGPT and Claude to connect directly to your account. You can create tests, add questions, and manage your content through an AI conversation — no need to open the app.Have a feature request or found something broken? Let us know.