mols-markdown-scripts

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the 'rumdl' CLI tool to perform Markdown formatting and link validation.
  • Evidence: Found in 'scripts/format_markdown.py' and 'scripts/validate_links.py'. The implementation uses the 'subprocess.run' method with argument lists, which is a secure way to execute external commands without shell interpolation risks.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided Markdown files, which creates an attack surface for indirect prompt injection.
  • Ingestion points: Markdown file content is read and processed in 'scripts/generate_index.py', 'scripts/validate_frontmatter.py', 'scripts/validate_headers.py', and 'scripts/markdown_size_guard.py'.
  • Boundary markers: The instructions do not explicitly include boundary markers or warnings for the agent when reading external file content.
  • Capability inventory: The skill has the capability to read from the filesystem, write index files, and execute formatting tools.
  • Sanitization: Content parsing is performed using established libraries ('pyromark', 'PyYAML'), and the scripts specifically use 'yaml.safe_load()' to prevent unsafe deserialization.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 02:43 PM
Security Audit — agent-trust-hub — mols-markdown-scripts