ClawTeam Multi-Agent Coordination

Warn

Audited by Socket on May 2, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

SUSPICIOUS: the skill is internally aligned with multi-agent coordination and does not show obvious credential theft or off-platform exfiltration, but it enables high-risk autonomous behavior by default. The main concern is broad, approval-skipping permissions for spawned sub-agents, which is disproportionate for a coordination skill and raises the operational security risk substantially.

Confidence: 85%Severity: 74%
Audit Metadata
Analyzed At
May 2, 2026, 08:28 AM
Package URL
pkg:socket/skills-sh/win4r%2FClawTeam-OpenClaw%2Fclawteam-multi-agent-coordination%2F@15cd957b2cbe3c670f3e28c01469d9e6007783a1