Claude SDK Expert
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional, providing guidance and Python/TypeScript code snippets for building autonomous agents. It focuses on documenting capabilities like computer use, tool calling, and Model Context Protocol (MCP) integration.
- [SAFE]: The content includes a dedicated security section that provides examples of defensive measures, such as input sanitization to block dangerous commands (including fork bombs and recursive deletions) and restricting file tool access to specific safe directories.
- [SAFE]: All external URLs and package references target official documentation, repositories, and services from trusted organizations (Anthropic) and well-known technology platforms (Model Context Protocol).
Audit Metadata