agents-claude-creator
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation and template library for building Claude Code subagents. A thorough review of the markdown instructions and asset files revealed no malicious intent, prompt injections, or persistence mechanisms.
- [SAFE]: Security best practices are a core focus of the instruction set. The documentation explicitly warns against over-privileged configurations, recommending the use of specific tool allowlists (e.g.,
Read,Grep,Glob) rather than inheriting all system tools by default. - [SAFE]: The example agents provided, such as the
code-reviewer.md, include explicit instructions for the subagent to check for exposed secrets, API keys, and credentials, promoting defensive coding practices. - [SAFE]: Technical references to platform features, such as
permissionMode, lifecyclehooks, andmcpServers, are documented correctly according to standard agent platform specifications without attempting to exploit these mechanisms.
Audit Metadata