judea-pearl

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of markdown instructions and a large corpus of academic reference data in JSON format. No malicious patterns were identified across any of the 77 analyzed files.
  • [PROMPT_INJECTION]: Instructions are entirely professional and academic. They provide clear guidelines for causal reasoning without attempting to override the agent's underlying safety protocols or system instructions.
  • [DATA_EXPOSURE]: There is no evidence of hardcoded credentials, access to sensitive local file paths, or unauthorized data collection. External URLs are directed to legitimate academic sources, repositories, and reputable websites.
  • [REMOTE_CODE_EXECUTION]: The skill does not contain any code or commands that perform external downloads or runtime execution of remote scripts.
  • [OBFUSCATION]: No malicious obfuscation, hidden characters, or deceptive encoding was found. Encoded data found in some URL parameters (e.g., in _workspace/discovery/interviews.json) was determined to be benign metadata from the source platforms used during the skill's generation process.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 05:30 AM