The Agent Skills Directory

[DATA_EXFILTRATION]: Performs network requests to fetch website content and consults pre-built scraper lists via Bright Data's official API endpoints. It follows security best practices by utilizing environment variables for API keys and configuration.
[COMMAND_EXECUTION]: Generates and suggests the execution of Python or Node.js scripts to validate scraping logic and ensure data quality, which is consistent with the skill's primary intended purpose.
[PROMPT_INJECTION]: Exposes an attack surface for indirect prompt injection by ingesting and analyzing untrusted HTML and JSON content from external websites.
Ingestion points: Site data is retrieved from external URLs using Web Unlocker and Browser APIs as described in SKILL.md and references/site-analysis-guide.md.
Boundary markers: There are no explicit markers or instructions used to isolate processed web content from the agent's internal logic.
Capability inventory: The skill possesses capabilities for file writing, network communication, and script execution.
Sanitization: No sanitization or filtering is performed on the fetched external data before the agent performs structural analysis.

scraper-builder