writing-skills

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a Node.js helper script at scripts/render-graphs.mjs that executes the local dot (Graphviz) command via execFileSync. This script is intended to render diagrams extracted from the SKILL.md of a target skill directory specified by the user.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and audit other agent skills, which establishes a surface for indirect prompt injection. Ingestion points: The skill reads SKILL.md files and associated reference documents from external skill directories. Boundary markers: The rendering script identifies input using standard Markdown code fences (e.g., ```dot). Capability inventory: The skill possesses file system read/write capabilities and the ability to execute the dot binary as a subprocess. Sanitization: The rendering script performs basic structural validation on the generated SVG output but includes an explicit warning that it does not sandbox the dot process or sanitize the final SVG for publication.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 12:03 AM