Skills
skills/acaprino/alfio-claude-plugins/agent-sdk-builder/Gen Agent Trust Hub

agent-sdk-builder

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill documents the official Claude Agent SDK, providing integration guides for TypeScript and Python.
  • [EXTERNAL_DOWNLOADS]: Mentions official packages from the Anthropic organization, including @anthropic-ai/claude-agent-sdk and claude-agent-sdk. These are well-known libraries from a trusted source.
  • [COMMAND_EXECUTION]: Describes the use of the Bash tool for executing shell commands, which is a core feature of the agent SDK. The documentation explicitly warns about the risks and provides remediation advice like using canUseTool callbacks to block destructive commands.
  • [SAFE]: Includes comprehensive security guidance (Section 16) advising on least-privilege tool access, budget caps, and sandbox isolation when executing untrusted code.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 03:14 AM