writing-skills

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script render-graphs.js uses child_process.execSync to run system commands (dot and which). This is used to render Graphviz diagrams extracted from markdown files into SVG format. While this involves subprocess spawning, it is a documented utility function restricted to local file processing.
  • [PROMPT_INJECTION]: The skill documents 'persuasion principles' (Authority, Commitment, Scarcity) intended to override an agent's natural 'rationalization' tendencies. By using imperative language and explicit anti-rationalization tables, the skill teaches how to create instructions that strongly influence agent behavior and bypass default decision-making logic in favor of strict protocol adherence (e.g., enforcing TDD).
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 02:13 PM
Security Audit — agent-trust-hub — writing-skills