claude-api
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation repository for the Claude API and Anthropic SDKs, including both standard API usage and the Agent SDK for building autonomous agents.
- [SAFE]: Remote resources and dependencies (e.g.,
pip install anthropic,npm install @anthropic-ai/sdk, and GitHub URLs) exclusively target official Anthropic and community-trusted repositories. - [SAFE]: All command execution examples (e.g., shell commands, MCP server setup) are instructional, standard for developer tools, and explicitly include safety warnings regarding human-in-the-loop review for destructive operations.
- [SAFE]: Code execution patterns described relate to Anthropic's server-side sandboxed environment, which is an official platform feature with built-in security controls.
- [SAFE]: The permission system documentation for the Agent SDK correctly emphasizes the use of 'default' modes and warns against skipping permissions without user intent.
Audit Metadata