The Agent Skills Directory

[SAFE]: The skill acts as a developer guide and utility suite for creating HTML reports within a specific platform. It does not contain instructions to bypass security filters or exfiltrate data. The provided scripts are for linting content against security policies.\n- [DYNAMIC_EXECUTION]: While the skill assists in generating HTML documents that include JavaScript, the documentation and implementation details specify that this code executes within a sandboxed iframe using a restrictive Content-Security-Policy (CSP) that blocks most browser APIs and network access.\n- [COMMAND_EXECUTION]: The skill includes a Python validation script (scripts/validate_page.py). This script uses standard libraries to perform static analysis on HTML files and does not execute the code it lints. It represents a security control rather than a threat.\n- [DATA_EXFILTRATION]: The skill describes a sanctioned postMessage protocol for communicating with the host platform (e.g., for file exports or opening links). The documentation explicitly notes that the host verifies the message source and applies size and type caps, aligning with secure inter-frame communication practices.

datalens-html-pages