The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches content from external URLs provided in the url parameter. This content is processed by the agent, creating a surface for indirect prompt injection if the source website contains instructions designed to manipulate the agent's behavior.
Ingestion points: Scraped content from the target URL via curl or fetch calls described in SKILL.md.
Boundary markers: None. The skill instructions specify returning raw response bodies directly to the agent.
Capability inventory: The agent has access to Bash, Read, Write, Edit, and Grep tools.
Sanitization: No sanitization of the scraped content is mentioned before it is returned to the agent context.
[COMMAND_EXECUTION]: The skill uses Bash to execute curl and node commands to perform scraping tasks and configuration reading.
[DYNAMIC_EXECUTION]: The skill utilizes node -e to dynamically execute Javascript code snippets for reading the API key from a local configuration file and for constructing/parsing API requests.

xcrawl-scrape