candid-improve
Warn
Audited by Snyk on Aug 18, 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). In SKILL.md Step 3 (source resolution) and Step 6 (refine loop), the agent ingests arbitrary free text from
--file <path>and/or inline text and/or the most recent substantive session output without any requirement to first select a specific trusted item.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata