syndly-sources

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow includes an “Add/Refresh” path where the agent accepts a runtime-supplied url/source, calls add_source or refresh_source, and Syndly will fetch outsider-authored web content (public pages/RSS/sitemaps) and then use it as context for the agent to report results, creating an indirect prompt-injection exposure risk.