agent-sdk-quickstart
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: The skill provides documentation and code examples for using the official Claude Agent SDK. All examples follow safety guidelines, including warnings against bypassing permissions and recommendations for using auditing hooks.
- [EXTERNAL_DOWNLOADS]: The skill references the installation of "@anthropic-ai/claude-agent-sdk" and "claude-agent-sdk" from official package registries. These are provided by a well-known and trusted organization.
- [SAFE]: No malicious patterns, such as prompt injection, data exfiltration, or obfuscation, were detected in the skill content.
Audit Metadata