babysit-pr

The skill is internally coherent for GitHub PR monitoring and remediation, with official/expected tooling and no obvious credential exfiltration path. The main risk is its powerful autonomous behavior: it can post publicly and mutate code over an extended loop while consuming untrusted external content, so it is best classified as a legitimate but high-impact skill rather than malware.