agent-collaboration
Warn
Audited by Socket on Apr 12, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the stated collaboration purpose matches the pane-control capabilities, and no external credential routing or obvious exfiltration is described. However, the core dependency on unverified tmux-bridge/smux tooling creates a material trust gap, and the skill can inject prompts/keystrokes into other panes and spawn additional agent instances.
Confidence: 80%Severity: 72%
Audit Metadata