The Agent Skills Directory

[SAFE]: The skill implements a mandatory 'Direction confirmation' step (Step 3) that requires the user to approve the gathered source materials and planned structure before the writing process begins, ensuring human oversight.
[SAFE]: Robust content filtering is defined in the workflow to automatically reject requests involving sensitive topics such as politics, military operations, religion, or controversial social issues.
[SAFE]: HTML generation uses the Jinja2 templating engine with auto-escaping enabled, protecting the resulting briefings from potential cross-site scripting (XSS) if malicious content were ingested from web sources.
[COMMAND_EXECUTION]: The script scripts/render.py utilizes a subprocess call to open the generated HTML file in a browser. This is implemented safely using a list of arguments without shell invocation (shell=False), which prevents command injection vulnerabilities.
[SAFE]: External dependencies are limited to jinja2, a well-known and trusted library for templating, and are installed using standard, safe practices.

topic-brief