The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches and processes content from external websites to discover email addresses, which represents an indirect prompt injection surface.
Ingestion points: The scrapeDomainForEmails function in references/full-guide.md performs HTTP GET requests to arbitrary user-provided domains and subpaths like /contact and /about.
Boundary markers: No specific delimiters or "ignore instructions" warnings are applied to the fetched HTML content before processing.
Capability inventory: The skill does not possess high-risk capabilities like shell execution or file-writing based on the scraped data; it primarily uses regex to extract strings.
Sanitization: The implementation uses a strict regular expression /[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}/g to extract only email-formatted strings, which effectively filters out most malicious command payloads.
[EXTERNAL_DOWNLOADS]: The skill documentation describes how to interact with several well-known lead generation and enrichment services.
The findEmailsForDomain function in references/full-guide.md makes outgoing network requests to Hunter.io, Apollo.io, Snov.io, and Clearbit.
These are established third-party services used for their intended purpose within the skill.
[COMMAND_EXECUTION]: The README.md provides a standard installation command using bunx to add the skill from the vendor's repository.

email-finder