product-changelog
Warn
Audited by Snyk on Jul 23, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.70). The required workflow in SKILL.md instructs using
belt app run ...with an expliciturlfield (infsh/agent-browser) which causes runtime browser-fetched public web page text (outsider-authored free text) to be read by the agent’s LLM context (indirect prompt injection path).
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata