The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and process data from external websites, which presents an attack surface for indirect prompt injection.
Ingestion points: Website content is retrieved via the Tavily API and returned in the results array (specifically raw_content and chunks) as described in SKILL.md.
Boundary markers: The skill documentation does not suggest using specific delimiters or "ignore instructions" wrappers when the agent processes this external data.
Capability inventory: The skill uses curl to fetch data, but the surrounding agent system may possess broader capabilities that could be targeted by instructions hidden in crawled websites.
Sanitization: There is no evidence of automated sanitization or filtering of the content before it enters the agent's context.
[COMMAND_EXECUTION]: The skill utilizes the curl command-line tool to perform HTTP requests to the Tavily API.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to api.tavily.com. This is an interaction with a well-known service for web crawling and search data.

tavily-crawl