beautiful-mermaid-ascii
Warn
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- EXTERNAL_DOWNLOADS (MEDIUM): The documentation in SKILL.md states that the tool automatically installs the beautiful-mermaid package from npm at runtime. As the package and author (lukilabs) are not part of the trusted organization list, this constitutes an unverifiable dependency installation that executes code outside of the skill's own directory.
- COMMAND_EXECUTION (LOW): The skill provides an installation script (scripts/install-mermaid-ascii) that symlinks files into ~/.local/bin. This modifies the user's execution environment and persistence in the shell path.
- PROMPT_INJECTION (LOW): The skill is designed to ingest and process untrusted Markdown content to extract and render Mermaid diagrams. 1. Ingestion points: Markdown files and standard input (stdin). 2. Boundary markers: The skill identifies content within triple-backtick fences specifically tagged with mermaid. 3. Capability inventory: The documented scripts perform file system reads and execute Node.js code from external dependencies. 4. Sanitization: No sanitization logic is visible as the core script files (scripts/mermaid-ascii and scripts/mermaid-ascii.mjs) were not provided for analysis.
Audit Metadata