The Agent Skills Directory

[SAFE]: The skill provides a controlled interface for inter-agent communication. Analysis of the source code confirms it acts as a thin wrapper for a host bridge (rlm.host_request), where critical security logic—including identity verification, access control, and rate limiting—is enforced by a system daemon rather than the skill itself. This architecture prevents sender identity spoofing and unauthorized messaging outside the agent's defined hierarchy.
[SAFE]: No unauthorized network operations, file system access, or sensitive data exposure were detected. The skill correctly validates inputs (e.g., receiver roles and names) before dispatching messages and does not handle or store credentials.
[SAFE]: No evidence of prompt injection, obfuscation, dynamic code execution, or persistence mechanisms was found. The skill's implementation is transparent and consistent with its stated purpose of managing session-to-session messaging.

agent-message