constitution
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is a pure markdown document providing behavioral guidelines. It contains no executable scripts, shell commands, or binary files.
- [SAFE]: No network operations, data exfiltration patterns, or obfuscation techniques were detected.
- [PERSISTENCE]: The 'Over time' section instructs the agent to persist user preferences in platform-standard configuration files such as
CLAUDE.mdandAGENTS.md. This is a documented feature for agent personalization and does not constitute a malicious persistence mechanism like shell profile modification. - [REMOTE_CODE_EXECUTION]: The skill mentions a
delegatetool but does not include any commands to download or execute external code. It merely refers to the tool as part of its procedural logic.
Audit Metadata