catch-up

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the agent to run Glean queries and chat for “important things happened [time period]” and to extract meetings/mentions, which at runtime will incorporate outsider-authored content from meetings/messages/issues stored in Glean into the LLM context via the Glean AI synthesis and retrieved items.