orchestrating-chat-agents
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a legitimate and well-structured library for programmatic API orchestration, supporting parallel processing, conversation threads, and prompt caching.
- [SAFE]: Credential management is handled securely by reading from the project's local file system (/mnt/project) rather than hardcoding secrets or using methods that expose keys to the agent's context.
- [SAFE]: The code relies solely on the official and well-known
anthropicPython SDK. - [SAFE]: The documentation includes explicit security guidance, such as the "Non-negotiable on all surfaces" rule which mandates manual review of delegated outputs to prevent the propagation of errors or malicious content.
- [SAFE]: No obfuscation, persistence mechanisms, or unauthorized privilege escalation patterns were detected in the source code or instructions.
Audit Metadata