The CamelHive Standard · v3.33.0
CH-SEC-051v3.33.0The output of an AI model is never executed or rendered as trusted code
Pinned as it read in catalog v3.33.0 — the current version of this rule may read differently, see CH-SEC-051.
- Area
- SEC
- Severity
- high
- Detection
- llm
- Corpus coverage
- not measured by the corpus