SECUpdates

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

• Third-party content exposure detected (high risk: 1.00). The skill's Update workflow explicitly launches parallel WebFetch requests to public sites (e.g., https://tldrsec.com, https://no.security, https://krebsonsecurity.com, https://thehackernews.com, https://schneier.com as listed in Workflows/Update.md and SKILL.md) and instructs the agent to extract, interpret, categorize, and score article headlines/summaries, so untrusted third-party content is read and can directly influence the agent's decisions and outputs.