The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill performs network operations to fetch content from a wide variety of advertising-related URLs provided in its reference files. This is used to verify the recency of publications. The list of targets includes well-known official platform blogs (such as those from Google, Meta, LinkedIn, Microsoft, and TikTok), major industry trade publications, and individual practitioner sites.
[INDIRECT_PROMPT_INJECTION]: By instructing the agent to fetch and process content from external websites to identify the most recent dated items, the skill introduces an indirect prompt injection surface. A malicious or compromised website could theoretically include text designed to influence the agent's behavior during this extraction process.
Ingestion points: The skill instructions in SKILL.md define a workflow to fetch canonical URLs and identify dated items (e.g., blog posts, podcast episodes).
Boundary markers: There are no specific delimiters or "ignore instructions" warnings implemented for the content fetched from the web.
Capability inventory: The skill's capabilities are limited to network fetching and text analysis; it does not include instructions for writing to the filesystem or executing shell commands based on the external data.
Sanitization: The skill lacks explicit sanitization or validation logic for the external text before it is processed by the agent to determine the date.

advertising-radar