runaway-guard
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and contains no executable code or scripts. It establishes safety protocols for the AI agent to follow when generating code that interacts with billed API services.
- [SAFE]: No evidence of prompt injection or malicious behavior was found. The instructions reinforce safety by requiring strict iteration bounds and cost limits on API calls.
- [SAFE]: All external references to SDKs (such as OpenAI, Anthropic, and Fal.ai) are for well-known, legitimate technology services and are used contextually to describe cost-management strategies.
- [SAFE]: No credentials, sensitive file paths, or network exfiltration patterns were detected. The skill specifically advises on the secure management of API keys and budget thresholds on provider platforms.
Audit Metadata