The Agent Skills Directory

Indirect Prompt Injection (LOW): The skill's primary purpose is to ingest and summarize untrusted content from the web, which is a classic vector for indirect prompt injection.
Ingestion points: Data enters the agent via start_crawl and subsequent processing of the downloaded site content.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands within the crawled content are mentioned in the workflow.
Capability inventory: The skill has the capability to read full site trees and report summarized results to the user/agent.
Sanitization: There is no evidence of sanitization or filtering of the HTML/text content before it is processed by the agent.
Data Exposure & Exfiltration (SAFE): While the skill documentation mentions customCookies and customHeaders for authentication, no hardcoded secrets or exfiltration patterns were detected. These are presented as configuration options for the user.

crawl-site