The Agent Skills Directory

[COMMAND_EXECUTION]: The skill extensively uses the webcmd CLI tool to list available data sources and execute search queries.
[EXTERNAL_DOWNLOADS]: The skill fetches data from external websites (such as Reddit, Twitter, and Wikipedia) and AI services (Grok, ChatGPT, Gemini) via the webcmd interface.
[PROMPT_INJECTION]: The skill processes content from untrusted external sources, which creates a potential surface for indirect prompt injection.
Ingestion points: External data is ingested through search results returned by webcmd calls across multiple categories (news, social media, shopping, etc.).
Boundary markers: The skill lacks explicit instructions for the agent to use delimiters or ignore instructions that might be embedded within the retrieved search results.
Capability inventory: The skill possesses command execution and network access capabilities through the webcmd tool.
Sanitization: The instructions do not describe any sanitization, filtering, or validation of the content retrieved from external sources.

smart-search