surf

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

• Third-party content exposure detected (low risk: 0.10). The workflow involves executing a CLI tool (surf) to query cryptocurrency data, prices, on-chain SQL, and news from an external API (https://api.asksurf.ai/gateway/v1), which returns data that the agent consumes and presents to the user. This constitutes consuming arbitrary API responses and web/news content.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to install and update a CLI tool via npx skills check asksurf-ai/surf-skills --skill surf, fetching code and instructions from an external repository without an enforced digest or immutable identifier, establishing an unverifiable runtime external dependency.