The Agent Skills Directory

[SAFE]: The skill navigates to scholar.google.com, which is a well-known technology service. This interaction is legitimate and necessary for the skill's stated purpose of academic search.- [SAFE]: Uses evaluate_script to run a scraper in the browser context. The JavaScript code is statically defined in the skill and is focused on extracting structured paper data from the DOM.- [PROMPT_INJECTION]: The skill processes untrusted web content, creating a surface for indirect prompt injection.
Ingestion points: Data is scraped from titles, snippets, and author fields in SKILL.md using CSS selectors like .gs_rt and .gs_rs.
Boundary markers: No delimiters or instructions to ignore embedded content are used in the reporting phase.
Capability inventory: The skill uses browser navigation and script execution via mcp__chrome-devtools.
Sanitization: Scraped text is presented directly to the agent without escaping or validation. This represents a standard scraping surface with low risk given the context.

gs-search