The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses bash -c to execute complex command chains involving curl and jq. This is used as a workaround for a documented platform limitation regarding environment variable persistence during pipes.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted content from external search engines (Google, Bing, YouTube, etc.) through the SerpApi service.
Ingestion points: Search results, snippets, and titles are retrieved from serpapi.com and piped into the agent's context in SKILL.md.
Boundary markers: Absent. The skill does not provide clear delimiters or instructions to the agent to treat the scraped content as untrusted data or to ignore embedded instructions.
Capability inventory: The skill relies on shell command execution via curl and jq for its primary functionality.
Sanitization: Absent. The raw text from the API response is filtered by jq but otherwise presented to the agent without escaping or validation.

serpapi