datastar

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow reads free text authored by the user from incoming Datastar write requests as “signals” via read_signals(request) (Python) / datastar.ReadSignals(r, &signals) (Go), which parses the request query/body that can be posted by an outsider.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's recommended script tag will fetch and execute remote JavaScript at runtime from https://cdn.jsdelivr.net/gh/starfederation/datastar@v1.0.2/bundles/datastar.js, so the agent depends on externally-hosted code that runs in clients.