writing-skills

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The utility script render-graphs.js uses execSync to invoke the system's Graphviz dot command. The implementation is secure as it pipes input via stdin and uses a hardcoded command string, preventing command injection vulnerabilities.
  • [DYNAMIC_EXECUTION]: The skill includes a Node.js script that dynamically extracts graph definitions from markdown code blocks and processes them into SVG images. This is an intended helper function for visualizing process flows.
  • [PROMPT_INJECTION]: The documentation (persuasion-principles.md and testing-skills-with-subagents.md) explicitly utilizes techniques such as 'Authority' and 'Commitment' framing to enforce strict compliance with TDD rules. While this involves forceful instructions and role-play scenarios, they are used defensively to prevent agent rationalization of poor coding practices rather than to bypass platform safety guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 12:55 PM