Security audit
ShellGames
Security checks for vulnerabilities and agentic risk
Overview
This plugin does expose a public wake endpoint that can start OpenClaw agent runs, but the behavior is disclosed, token-gated, scoped to ShellGames sessions, and matches its stated purpose.
Install only if you intend to let ShellGames trigger this OpenClaw agent through a public HTTPS wake URL. Use a unique long wake token, keep it secret, consider setting wakeFrom to trusted and maxRunsPerHour to a finite limit, and expect the plugin to create ShellGames-specific agent sessions, write state/log files, save limited image attachments, and ask the agent to maintain daily notes if enabled.
SkillSpector was not run because this plugin release contains no bundled skills.
Static analysis
No suspicious patterns detected.