The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes mkdocs and git commands using the subprocess.run method with argument lists. By avoiding shell=True, the implementation prevents potential shell injection attacks from malicious user input in fields like commit messages or branch names.
[EXTERNAL_DOWNLOADS]: The skill references standard dependencies (mkdocs, mkdocs-material, pyyaml) which are widely used and trusted in the software industry for building static documentation sites.
[PROMPT_INJECTION]: The core instructions in SKILL.md are focused on documentation frameworks and do not contain any instructions that attempt to bypass AI safety guardrails or reveal system prompts.
[DATA_EXFILTRATION]: Technical review found no evidence of credential harvesting or sensitive file access. The skill's file operations are scoped to project documentation directories, and network access is managed through standard Git operations.

documentation-writing