The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the execution of an external CLI tool named grokipedia for all its knowledge retrieval, search, and editing functionality.
[INDIRECT_PROMPT_INJECTION]: The skill provides the agent with capabilities to fetch content from an external, unofficial wiki site, creating a surface for indirect prompt injection attacks.
Ingestion points: Untrusted content is ingested into the agent's context through grokipedia search, grokipedia page, grokipedia preview, and grokipedia list commands.
Boundary markers: The skill instructions lack explicit delimiters or instructions for the agent to ignore any embedded commands or directions found within the retrieved wiki articles.
Capability inventory: The skill allows for subsequent actions based on the ingested data, including further searches and the preparation of contribution/edit requests (grokipedia suggest, grokipedia edit).
Sanitization: There is no evidence of content sanitization or validation to prevent the agent from acting on malicious instructions hidden within wiki articles.

grokipedia-cli