smartclaws-reader

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (SKILL.md "Reading Data") instructs running "smartclaws read --channel " against a public SKALE RPC URL to fetch on-chain channel messages, which are arbitrary producer/user-generated data that the agent parses and uses to compute averages, thresholds, and trends, so untrusted third-party content can influence its actions.