using-spacy-nlp

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). In scripts/serve_model.py, the FastAPI runtime ingests outsider-authored free text directly from HTTP POST bodies (/classify and /classify/batch) via input.text / input.texts, then passes it to the spaCy model (doc = nlp(input.text) / for doc in nlp.pipe(input.texts, ...)).