The Agent Skills Directory

[SAFE]: The skill provides explicit security guidance for managing credentials, instructing users to use environment variables or headers rather than hardcoding tokens into scripts or URLs.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data scraped from external social media profiles and third-party websites during the email discovery process.
Ingestion points: Data is ingested via the flash_scraper/creator-leads-scraper Actor, which retrieves bio text and crawls external creator websites.
Boundary markers: None explicitly defined in the prompt instructions, although the output is constrained to a structured 51-column dataset schema.
Capability inventory: Uses apify actors call to execute scrapers and apify datasets get-items to retrieve results.
Sanitization: No specific sanitization or filtering of the scraped text (bio/website content) is mentioned before it is returned to the user or processed by the agent.

apify-creator-emails