lov-version-management-manual
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely within the local environment, reading project configuration and repository status (diffs) to generate changesets. There are no network operations or external data exfiltration patterns.
- [SAFE]: The
profile_store.pyscript implements a proactive security filter (SENSITIVE_PARTS) that prevents the persistence of sensitive data like API keys, tokens, or passwords in the shared profile context. - [SAFE]: The skill uses atomic file writing and strict permission settings (
0o600) for profile data, ensuring that configuration files are handled securely and remain accessible only to the owner. - [SAFE]: Technical implementation follows security best practices, including the use of
yaml.safe_load()in the validation scripts to mitigate risks associated with unsafe deserialization of configuration files. - [SAFE]: The execution boundaries in
SKILL.mdemphasize that the agent should remain read-only during analysis and only perform modifications after explicit user review and authorization.
Audit Metadata