claude-agent-sdk-expert
Pass
Audited by Gen Agent Trust Hub on Mar 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill provides a shell script 'scripts/scan-agent-patterns.sh' designed to perform static analysis on local source code directories. The script utilizes standard utilities (grep, sed, rg) to identify anti-patterns such as missing iteration limits or unsafe JSON parsing in agent code.
- [SAFE]: The core of the skill is composed of architectural guides and best practice documentation covering agentic loops, tool design, and error handling. These files are instructional and do not contain executable malicious code.
- [EXTERNAL_DOWNLOADS]: The documentation references several official Model Context Protocol (MCP) servers (e.g., '@modelcontextprotocol/server-github') as examples for integration. These are well-known, community-standard packages used for extending agent capabilities.
Audit Metadata