skills/thatjuan/agent-skills/batch-implement/Socket

batch-implement

Warn

Audited by Socket on Sep 9, 2026

1 alert found:

Security

SecuritySKILL.md

MEDIUM

SecurityMEDIUM

SKILL.md

SUSPICIOUS: the skill's core behavior matches its stated purpose, and its external tooling/data flows are mostly coherent and official. The main risk is scope and autonomy: it authorizes unattended code changes and PR creation with minimal validation and reduced visibility into subagent behavior.

Confidence: 90%Severity: 76%

Audit Metadata

Analyzed At

Sep 9, 2026, 07:17 PM

Package URL

pkg:socket/skills-sh/thatjuan%2Fagent-skills%2Fbatch-implement%2F@1eab0a499202457f1363b84c294fba201f91566b5ee0338a60d4507b22bc6f07