documentation-writing
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
subprocessmodule to automate documentation workflows.github_pages/generator.pyexecutesmkdocs buildto generate static sites, andgithub_pages/deployer.pyexecutes multiplegitcommands for deployment. To mitigate command injection risks, the skill implements specific validation functions like_validate_branch_name(which blocks dangerous characters such as semicolons and pipes) and_validate_github_url(which ensures operations are restricted to GitHub repositories). - [EXTERNAL_DOWNLOADS]: The
github_pages/README.mdfile recommends the installation ofmkdocs,mkdocs-material, andpyyaml. These are well-known and trusted packages within the Python ecosystem for documentation purposes and are categorized as safe dependencies. - [INDIRECT_PROMPT_INJECTION]: The skill has a defined attack surface for indirect prompt injection as it ingests content from project files and user instructions to generate documentation.
- Ingestion points: Content is read from the
docs/directory and projectREADME.mdfiles during the generation and validation phases. - Boundary markers: The skill does not explicitly use security delimiters for user-provided text, but it enforces a strict structural framework (Diataxis) that provides implicit context separation.
- Capability inventory: The skill can write to the local file system and execute git and mkdocs CLI tools.
- Sanitization: Command-line arguments derived from user input (like branch names and URLs) are sanitized using dedicated validation routines to prevent injection attacks.
Audit Metadata