JFrog Patterns

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's workflows explicitly create remote repositories that point to public registries (e.g., repositories.md uses "https://registry.npmjs.org"), enable curation/Xray scans, and instruct the agent to run/review curation audits and Xray violations (flow-suggestions, supply-chain-security, multi-site), so untrusted third-party package content and metadata from public registries are fetched and used to gate/promote actions.