cloud-claw
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious prompt injection, data exfiltration, or obfuscation patterns were identified in the skill instructions or configuration.
- [COMMAND_EXECUTION]: The skill utilizes local 'altllm cloud-claw-*' commands to perform its functions. These commands are consistent with the skill's stated purpose of managing VM lifecycles and infrastructure.
- [EXTERNAL_DOWNLOADS]: The skill references source code and documentation on GitHub (alt-research/cloud-claw). These are provided as static references for the agent to understand the underlying implementation and do not involve automated execution of remote scripts or unverified dependencies.
Audit Metadata