The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes an external script competitor-tracker.py with the save-baseline argument to persist collected data to the local filesystem.
[PROMPT_INJECTION]: The skill processes untrusted data from external sources (competitor websites, Google Ads Transparency Center, Meta Ad Library) which could contain hidden instructions targeting the agent.
Ingestion points: Competitor URLs provided by users and subsequent automated crawling of pricing, product, and ad library pages.
Boundary markers: None identified; external content is processed and synthesized directly into intelligence briefs.
Capability inventory: File system writes via competitor-tracker.py and external notifications via send-notification or email.
Sanitization: No explicit sanitization or validation of the ingested HTML/text content is described.

competitor-monitor