dalio-perspective

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is composed of static documentation and persona-driven instructions. It contains no executable code, scripts, or binaries.- [NO_CODE]: There are no functional code components (Python, Node.js, Shell) within the skill package. The 'dalio-perspective' persona is implemented purely through natural language instructions in SKILL.md.- [DATA_EXFILTRATION]: No network access or sensitive file system operations are defined. The skill does not attempt to read credentials, environment variables, or private user data.- [PROMPT_INJECTION]: The instructions focus on adopting a specific analytical persona ('Dalio Perspective'). While it encourages 'Radical Transparency' and 'Radical Truth', it does not contain bypass markers, jailbreak attempts, or instructions to ignore safety filters.- [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or dynamic context injections (e.g., !command) present in the skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 02:50 AM
Security Audit — agent-trust-hub — dalio-perspective