beautiful-mermaid-ascii

Warn

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The scripts/mermaid-ascii script performs a silent installation of the beautiful-mermaid package using npm install at runtime. The package is downloaded from the public npm registry into a local cache directory ($XDG_CACHE_HOME/beautiful-mermaid-ascii or /tmp/beautiful-mermaid-ascii).
  • [DYNAMIC_EXECUTION]: The scripts/mermaid-ascii.mjs script uses createRequire to resolve the location of the installed beautiful-mermaid library and then uses a dynamic import() call to load and execute the renderer function from a computed path.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from .mmd or .md files provided by the user or found in the environment.
  • Ingestion points: scripts/mermaid-ascii reads input from files specified as command-line arguments or via standard input.
  • Boundary markers: There are no specific delimiters or instructions provided to the agent to ignore potentially malicious embedded instructions within the diagrams.
  • Capability inventory: The skill executes shell commands and Node.js scripts to perform rendering and manages its own dependencies via npm.
  • Sanitization: The extractMermaidFences function in scripts/mermaid-ascii.mjs performs basic text extraction of Mermaid blocks but does not sanitize the contents of the diagrams before passing them to the third-party rendering library.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 14, 2026, 10:47 AM