blog-writing-specialist

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a library of writing guidelines and templates. It does not include executable scripts, network requests, or credential handling. All instructions are focused on content generation and stylistic adherence.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user data ('brain dumps') as its primary input. While this creates a surface for indirect prompt injection where a user could embed instructions to influence the agent's output, the skill lacks dangerous capabilities (like file system writes or network exfiltration) that would make such an injection high-risk. There are no specific boundary markers defined in the prompt to separate user data from instructions.
  • [COMMAND_EXECUTION]: The skill mentions using charting libraries like matplotlib or Chart.js for visualizations. While these often require code execution via a tool (like a Python interpreter), the skill itself provides instructions to 'avoid third-party execution risks' by using self-hosted or local tools, demonstrating a security-conscious approach to diagram generation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 12:37 PM
Security Audit — agent-trust-hub — blog-writing-specialist