work-with-pr
Warn
Audited by Socket on Mar 27, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The workflow is mostly aligned to its stated PR-lifecycle purpose and uses official GitHub/git/Bun tooling, but it grants an AI agent broad autonomous repository powers: unbounded fix/push loops, processing untrusted external review content, transitive skill loading, and automatic merge/delete actions. Main risk is high operational autonomy and prompt-injection exposure, not confirmed malware or credential theft.
Confidence: 89%Severity: 76%
Audit Metadata