system-context
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious behavior detected. The skill is a documentation-based workflow guiding the agent to use specific Red Hat Lightspeed MCP tools for inventory analysis.
- [PROMPT_INJECTION]: The use of emphatic markers such as 'CRITICAL' and 'IMPORTANT' in the description are functional instructions for tool orchestration (directing the agent to use this skill instead of raw tools) and do not represent attempts to bypass safety filters or jailbreak the AI.
- [DATA_EXFILTRATION]: While the skill involves reading sensitive system information (IP addresses, packages, processes), this is performed via authorized internal MCP tools for administrative purposes. There are no patterns suggesting data is being sent to unauthorized external domains.
- [SAFE]: The skill processes external system data (tags, hostnames), which is a common surface for indirect prompt injection. However, the skill lacks high-risk capabilities like arbitrary code execution or external network requests, and the data is used strictly for classification and strategy synthesis.
Audit Metadata