robius-state-management
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Data Exposure & Exfiltration] (SAFE): The skill documents standard local data persistence patterns. It uses legitimate application directories via the
dirscrate (e.g.,dirs::config_dir()) and does not attempt to access sensitive system files like SSH keys or cloud provider credentials. - [Persistence Mechanisms] (SAFE): The skill provides templates for saving application state and user preferences between sessions. These are legitimate application features and do not utilize malicious persistence techniques such as modifying system shell profiles or creating unauthorized background services.
- [Credential Exposure] (SAFE): While the patterns include structures for handling session data (e.g.,
ClientSessionPersistedwith apassphrasefield), they are provided as architectural templates. No hardcoded API keys, tokens, or actual credentials were found in the source code. - [Remote Code Execution] (SAFE): The skill does not contain any instructions for downloading and executing remote scripts or binary files. The mentioned network operations (via
MolyClientormatrix_sdk) are standard for the described application use cases. - [Indirect Prompt Injection] (SAFE): Although the skill describes patterns for reading from local files, the data is used for structured application state (UI themes, favorites, window geometry) and is not interpolated into LLM prompts, mitigating the risk of indirect injection.
Audit Metadata