siyuan
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill functions solely as a documentation resource for SiYuan Note. Analysis of the provided files confirms that all code snippets and instructions are intended for local note management. No malicious patterns, obfuscation, or unauthorized data transmission were detected.
- [COMMAND_EXECUTION]: The skill includes examples of using curl and the Python requests library to interact with the SiYuan Note REST API. These examples are for legitimate tasks such as creating, updating, or deleting notes on the user's local instance.
- [DATA_EXFILTRATION]: Documentation provides details on the SiYuan SQLite schema and SQL querying capabilities. While this involves accessing user data (notes), the skill does not contain any code for exfiltrating this data to remote servers; all data interactions are restricted to the local environment.
- [PROMPT_INJECTION]: The skill references the processing of user-generated content in notes and templates, which represents a potential indirect prompt injection surface. However, the documentation includes security-minded advice, such as recommending the use of SQL placeholders in templates to prevent injection attacks.
Audit Metadata