agentsop-cost-tiered-models
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate engineering patterns for cost-aware model selection and multi-agent orchestration. No malicious patterns, prompt injections, or persistence mechanisms were detected.
- [EXTERNAL_DOWNLOADS]: All external references target official documentation and reputable community discussions for well-known AI frameworks (aider.chat, github.com/stanfordnlp, docs.vllm.ai, langchain-ai.github.io). These are documented neutrally and do not introduce risk.
- [NO_CODE]: The skill consists of markdown instructions and metadata; it does not include executable scripts or remote code execution commands within its own definition.
Audit Metadata