The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required workflow calls Google Calendar `get_events` with `detailed: true` over a date range, which returns event titles, descriptions, and attendees, and any external sender can place an invite with attacker-written text onto the user's primary calendar without the agent selecting a specific item.
kmondlane/ExoMind
8794d26
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.