Responsive Agent
Security checks across malware telemetry and agentic risk
Overview
The skill's instructions, requirements, and behavior are coherent with its stated purpose (a responsiveness/spawn-subagent pattern) and it does not request unrelated credentials, installs, or access.
This skill is an architectural pattern (no installs or secrets) and appears coherent. Things to consider before enabling: (1) at runtime subagents will execute local commands and perform network I/O — ensure the agent's execution context and permissions are appropriately restricted and audited so spawned tasks can't access secrets or sensitive files; (2) test the queue/concurrency behavior in a controlled environment to avoid unexpected resource use (nested spawns and recursive tasks can escalate load); (3) note small metadata inconsistencies (SKILL.md version 1.8.0 vs _meta.json 1.3.0 and registry 1.8.1) — not a functional problem but worth confirming you have the intended version; (4) because this is instruction-only there is no code for third-party review — treat the runtime behavior as the security surface and limit agent privileges accordingly.
SkillSpector
SkillSpector findings are pending for this release.
VirusTotal
No VirusTotal findings