foragent-call

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow has the agent receive and act on messages from incoming ForAgent calls placed by other agents, which an outsider with the number can initiate without the agent selecting content first (human pickup approval gates the call but does not sanitize its text).

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill directs the user to obtain a setup message from https://foragent.io and paste it into the agent, which configures a plugin and installs a notifier, so that remote content is adopted as setup instructions; because the publisher is attributable to a single vendor's own site without enforced signature or digest verification, it is treated as common/reputable unofficial rather than strong first-party provenance, yielding medium.