The Agent Skills Directory

[PERSISTENCE]: The skill utilizes the cron tool to establish automated, recurring task execution (e.g., 0 9 * * *). While this is central to the skill's purpose as a social media scheduler, it functions as a persistence mechanism to maintain agent activity across sessions without direct user intervention.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external websites, which constitutes a potential attack surface for indirect prompt injection.
Ingestion points: The skill uses web_fetch to retrieve content from external news sources such as theverge.com, techcrunch.com, and news.ycombinator.com (documented in SKILL.md).
Boundary markers: Absent. The instructions do not specify any delimiters or instructions to the model to ignore potential directives embedded within the fetched research data.
Capability inventory: The skill can execute browser automation (including typing and clicking on x.com), write to local JSON files for logging and drafting, and spawn additional agent sessions.
Sanitization: There is no evidence of filtering, sanitization, or schema validation for the data scraped from external sources before it is processed by the generation pipeline.
[EXTERNAL_DOWNLOADS]: Fetches research data from well-known technology news platforms (The Verge, TechCrunch, and Hacker News).
[DYNAMIC_EXECUTION]: Uses the sessions_spawn capability to dynamically instantiate sub-agent sessions to handle content generation workflows in parallel.

social-media-agent