Heir Cognitive System Curation Skill
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [SAFE]: The skill's primary purpose is data sanitization and repository maintenance, specifically aimed at removing personal data and sensitive configurations from automated agent deployments.\n- [COMMAND_EXECUTION]: The skill includes instructions for using the
greputility to search for sensitive file references. This is used as a read-only validation step to ensure that the distribution payload is clean and contains no internal-only documentation.\n- [DATA_EXFILTRATION]: The skill instructions focus on mitigating data exposure by identifying and ensuring the exclusion of sensitive files such as user profiles (USER-PROFILE.md), personal configurations (user-profile.json), and private session memories from the repository distribution.
Audit Metadata