soros-perspective

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill consists entirely of documentation and instructional prompts in Markdown format. There are no scripts, binaries, or automated tasks included that could execute commands on the host system.
  • [PROMPT_INJECTION]: The instructions provide a persona and analytical methodology for the AI to follow when asked for a specific perspective. These prompts do not attempt to override safety guardrails, subvert core instructions, or extract system prompts.
  • [EXTERNAL_DOWNLOADS]: The installation instructions in the README reference npx skills add nuwa-skills/soros-skill. This uses the standard distribution tool for the platform, and the package name is consistent with the skill's author context (nuwa-skills).
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided market and policy scenarios. While it lacks explicit boundary markers for user input, it does not possess tools or capabilities (like file writing or network access) that would create a high-risk surface for indirect prompt injection attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 06:34 PM
Security Audit — agent-trust-hub — soros-perspective