The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch content from external web pages and search engines, which represents a surface for indirect prompt injection where malicious instructions could be embedded in the scraped data.
Ingestion points: Tools described in SKILL.md such as scrape_page, search_web, and search_amazon allow the ingestion of arbitrary web data.
Boundary markers: The skill does not provide specific instructions for using delimiters or protective prompts when the agent processes the fetched content.
Capability inventory: The skill facilitates network operations to the api.fastaf.co domain to retrieve web data.
Sanitization: No explicit sanitization, filtering, or validation of the retrieved external content is described.
[EXTERNAL_DOWNLOADS]: The documentation references the @x402/fetch Node.js package as an example client for implementing the x402 payment protocol.

fastaf-web-data