zr-check-in
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill contains defensive instructions in 'references/agent-integrations.md' that advise the agent to treat external content as data and scan for injection patterns (such as 'ignore previous instructions'). These are safety guidelines for the agent and not an attempt to subvert the LLM.
- [EXTERNAL_DOWNLOADS]: Fetches platform-specific updates via 'npx skills' and pulls the 'zenonred/probe' Docker image. These resources originate from the official vendor infrastructure and are consistent with the skill's purpose.
- [COMMAND_EXECUTION]: The skill uses various shell commands for environment preparation, wallet management, and service control. These operations are standard for onboarding and are performed at the user level.
- [REMOTE_CODE_EXECUTION]: Employs 'npx' to run vendor-provided management tools. This is the intended delivery mechanism for the platform's toolset.
- [SAFE]: All persistence mechanisms (systemd, launchd, cron) and credential management (local wallet password storage with restricted permissions) are transparently documented and follow standard security practices for CLI-based agent tools.
Audit Metadata