Explains prompt injection in graded content
https://knobit.org/id/knobit/T-353
Also as Turtle · JSON-LD · RDF/XML · AI benchmark on Skilirium
| Context functions | Object understanding |
|---|---|
| Knobit creation methods | AI generated |
| Knobit types | Knowledge |
| LCG purposes | Assessing the learner's knowledge or skill state |
| LCG purposes | Opening up a learning outcome |
| action verb | Explains en |
| assessment criterion | Answers a written question on the prompt injection correctly en |
| assessment method | Written question en |
| conforms to profile | https://oppekava.edu.ee/a/Knobitite_kirjeldamise_ja_seostamise_reeglid#LCG-KNOBIT-0.1 |
| dcterms:identifier | T-353 |
| dcterms:license | https://creativecommons.org/licenses/by/4.0/ |
| dcterms:source | Practice the stage 3 reference and review checklist rest on en |
| expected result | Text in content a model processes that tries to instruct it; a graded answer is data, never an instruction (OWASP Top 10 for LLM Applications 2025, LLM01). en |
| owl:sameAs | https://bitbucket.org/reffa/ontology/id/knobit/skill-assessment/03-answer-grading/T-353 |
| rdfs:label | Explains prompt injection in graded content en |
| schema:creativeWorkStatus | accepted |
| schema:isPartOf | Skill assessment: knobit registers, test items and answer grading |
| skos:definition | Text in content a model processes that tries to instruct it; a graded answer is data, never an instruction (OWASP Top 10 for LLM Applications 2025, LLM01). en |
| skos:prefLabel | Explains prompt injection in graded content en |
| target object | prompt injection en |
Used by
| applies knowledge | Recognises an instruction in a graded answer |
|---|---|
| applies knowledge | Reports an instruction found in a graded answer |
| includes knobit | Grades answers against a rubric criterion by criterion, as a calibrated grader would |
| includes knobit | Prompt injection |
| requires knobit | Grade answers against a rubric |