bt-panel-ops
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill outlines a legitimate operational procedure for server diagnostics using standard system commands and local configuration file paths defined in SKILL.md.
- [SAFE]: Robust security rules are integrated into the skill's core logic, specifically requiring explicit user consent for destructive operations and ensuring that no sensitive credentials or tokens are hardcoded or exfiltrated to external domains.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface through the processing of untrusted system and service log data.
- Ingestion points: The skill instructions in SKILL.md specify reading data from system and service logs (journalctl, docker logs, auth.log) and web server configuration files (vhost).
- Boundary markers: The 'Rules' and 'Workflow' sections mandate the use of structured output templates and require the agent to cite only specific, relevant evidence snippets rather than full untrusted content.
- Capability inventory: The skill enables the agent to utilize powerful system management tools such as systemctl, docker, and firewall-cmd, as well as database operations via a local MCP integration.
- Sanitization: The instructions explicitly require the desensitization of all credentials and personal data before reporting, and prohibit bypassing safety controls or executing scripts from unknown sources.
Audit Metadata