mkdocs-github-pages-deployment
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides scripts that automate the processing of markdown files (.md) in local directories (policies/, narratives/, procedures/). These scripts read file content, apply regex-based transformations, and write the modified content back to disk.
- Ingestion points: md_file.read_text() in multiple Python scripts.
- Boundary markers: Absent; scripts process entire file contents without distinguishing instructions from data.
- Capability inventory: File-write operations (write_text) and command execution (pip install, mkdocs build).
- Sanitization: Absent; content is processed via regex and string substitution without validation or escaping.
- [DYNAMIC_EXECUTION]: The skill contains several Python scripts for the agent to execute locally. These scripts perform file system operations and use regular expressions to modify local documentation files.
- [COMMAND_EXECUTION]: The skill includes shell commands for package installation and site building using the mkdocs toolset.
Audit Metadata