sota-sandboxing
Warn
Audited by Snyk on Jun 17, 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). In AUDIT/BUILD runtime, the workflow ingests outsider-authored free text such as issue/PR/wiki/discussion bodies or other external messages into the agent’s LLM context (prompt-injection vector described in
05§1.1 and §2.3), so the LLM can receive attacker-controlled prose.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata