kuroco-admin-mcp
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The documentation provides clear instructions for OAuth 2.0 authentication and authorization, emphasizing the use of minimal scopes (e.g.,
mcp:tools.read) and the implementation of IP-based access restrictions. - [SAFE]: Placeholders (e.g.,
<site_key>,<token>) are used throughout the command examples, ensuring that no real credentials or sensitive instance-specific information is hardcoded in the skill. - [SAFE]: The skill includes a dedicated
whoamitool designed to let users verify their effective permissions, connection scope, and site identity before performing administrative actions. - [SAFE]: File upload functionality is handled securely via a two-stage process using presigned URLs and temporary storage, which limits the exposure of the final destination storage.
- [SAFE]: The instructions for configuring MCP clients (like Claude Code or ChatGPT) follow standard protocols and direct users to the official vendor documentation for implementation details.
Audit Metadata