sales-devta
Warn
Audited by Snyk on Jun 14, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.75). Devta’s runtime workflow includes “Draft Posts” that “research[ ] using Google and Reddit Answers” and “scans trending topics,” so the agent can ingest outsider-authored web/forum text into the LLM context to generate drafts.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata