pr-watch
Warn
Audited by Socket on Mar 21, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill’s GitHub-only data flow is coherent with its PR-watching purpose, and install trust is relatively low risk because it uses the official gh CLI and GitHub APIs. The main issue is disproportionate autonomy: it continuously monitors untrusted PR comments and can decide to edit code, reply publicly, and possibly push changes in the background without per-action user confirmation.
Confidence: 89%Severity: 74%
Audit Metadata