The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill correctly utilizes the platform's create_scheduled_task tool to establish a recurring automation for fetching and summarizing news content.
[DATA_EXPOSURE]: The skill asks for the user's email address and preferred news topics. This data collection is transparent and necessary for the skill's stated purpose of delivering a personalized news digest.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes external data from news sources.
Ingestion points: External news headlines and content retrieved via search in SKILL.md.
Boundary markers: Absent in the task instructions provided to the agent.
Capability inventory: Uses create_scheduled_task in SKILL.md to perform search and email operations.
Sanitization: No explicit sanitization or filtering of news content is mentioned.

daily-news-digest