Rubric-based grading with feedback and integrity checks

The scenario grades the answer against your rubric, checks it for plagiarism, re-marks borderline cases, writes the student feedback, and logs everything to a sheet.

  • One consistent standard across the whole stack
  • A plagiarism check on every submission
  • Borderline cases re-marked, serious ones escalated to the teacher
  • Every student gets real feedback, not just a number
Set it up for yourselfScriptera

How it works

To move data from Microsoft SQL to botHelp automatically, use a ready-made scenario — no manual exports.

  1. Starts when: Manual Trigger
  2. Then: Workflow Configuration
  3. Then: Retrieve Student Answer and Rubric
  4. Then: Rubric Interpreter Agent
  5. Then: Primary Marker Agent
  6. Then: Quality Moderator Agent
  7. Then: Feedback Generator Agent
  8. Check: Route by Moderation Decision
  9. If yes: Secondary Marker Agent
  10. Check: Merge Marking Results
  11. If no: Prepare Escalation Data
  12. Then: Send Escalation Alert
  13. Check: Merge All Paths
  14. Then: Final Output Compilation
  15. Then: Log to Google Sheets
  16. If no: Calculate Statistics
  17. Check: Check Integrity Flags
  18. If yes: Plagiarism Analyzer Agent
  19. Then: Prepare Integrity Report
  20. Then: OpenAI Chat Model - Rubric
  21. Then: OpenAI Chat Model - Marker
  22. Then: OpenAI Chat Model - Moderator
  23. Then: OpenAI Chat Model - Feedback
  24. Then: Structured Output Parser - Rubric
  25. Then: Structured Output Parser - Marker
  26. Then: Structured Output Parser - Moderator
  27. Then: Structured Output Parser - Feedback
  28. Then: OpenAI Chat Model - Secondary
  29. Then: OpenAI Chat Model - Plagiarism
  30. Then: Structured Output Parser - Secondary
  31. Then: Structured Output Parser - Plagiarism

You can launch this Microsoft SQL + botHelp integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios