The Agent Skills Directory

[SAFE]: The skill processes local Markdown files to generate a static HTML visualization for knowledge bundles without performing network operations during execution.
[DATA_EXPOSURE]: The script uses pathlib's is_relative_to method to verify that all file read operations remain within the user-specified bundle directory, effectively preventing directory traversal attempts.
[DYNAMIC_EXECUTION]: The Python script utilizes yaml.safe_load for parsing concept metadata, which prevents the execution of arbitrary Python objects during the deserialization of YAML content.
[INDIRECT_PROMPT_INJECTION]: Although the skill handles untrusted Markdown content, the resulting HTML file uses DOMPurify to sanitize rendered Markdown before display, which provides a robust defense against Cross-Site Scripting (XSS) in the browser.
[EXTERNAL_DOWNLOADS]: The skill suggests installing pyyaml if the primary package manager is unavailable. pyyaml is a standard, reputable library in the Python ecosystem. The generated HTML also references well-known libraries from the jsdelivr.net CDN.
[OBFUSCATION]: A helper function is used to escape HTML special characters when embedding JSON data into script tags. This is a standard security practice to prevent script injection rather than a malicious attempt to hide code.

visualize