self-hosted-llm-batch-inference

Warn

Audited by Snyk on Nov 16, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (medium risk: 0.30). The required recurring workflow reads a corpus of input records from S3 shards (e.g., customer review text in record_id/input.text form) and feeds that text to the LLM without per-item selection, and such records are plausibly outsider-authored submissions that accumulate into the monitored corpus.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 16, 2026, 06:20 PM
Issues
1