The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides shell command examples for gathering data from external sources. For instance, in references/newsjacking.md, it includes curl commands to fetch RSS feeds from Google News and query the Hacker News (Algolia) API, as well as accessing Reddit's JSON API. These commands are used to automate the detection of trending news stories.
[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch content from various external domains, including news.google.com, hn.algolia.com, www.reddit.com, and various podcast hosting platforms to perform research and guest preparation.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a significant surface for indirect prompt injection because its primary function involves ingesting and processing untrusted data from the web.
Ingestion points: External news articles, RSS feeds, social media profiles (X, LinkedIn), and podcast transcripts (references/podcast-guest-prep.md, references/newsjacking.md).
Boundary markers: None identified. The instructions do not specify the use of delimiters or warnings to ignore instructions that might be embedded in the fetched external content.
Capability inventory: The skill uses dev-browser for web navigation and curl for API requests. It also performs file-write operations to maintain media lists (e.g., .agents/media-list.md).
Sanitization: No explicit sanitization or filtering of external content is mentioned before it is processed by the agent.

public-relations