talk-to
Warn
Audited by Snyk on Jun 21, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the LLM context via
arra_thread_read({ threadId })in Mode 3/4, because the agent’s thread contents (messages authored by the other party/agent) are read back and then used to compose follow-ups.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata