hoffman-perspective

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and referenced materials do not contain any malicious patterns, obfuscated code, or unauthorized data access commands. The skill's primary function is to simulate a specific persona based on publicly available business literature.
  • [EXTERNAL_DOWNLOADS]: The README provides a command to install the skill (npx skills add nuwa-skills/hoffman-skill) and links to official project repositories (github.com/nuwa-skills/*). These references are part of the standard distribution mechanism for the nuwa-skills ecosystem and target known vendor-owned resources.
  • [PROMPT_INJECTION]: The skill uses role-play instructions to define the Reid Hoffman persona. No attempts to bypass safety filters, override system prompts, or extract internal configuration were detected. The 'Agentic Protocol' defined in SKILL.md provides a structured way for the model to process queries within the intended scope.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 03:51 AM
Security Audit — agent-trust-hub — hoffman-perspective