kubesense-dashboards
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill is a legitimate technical guide for interacting with the KubeSense dashboard MCP server and generating valid JSON configurations.
- [INDIRECT_PROMPT_INJECTION]: The skill documents an attack surface where the agent ingests external telemetry metadata to build dashboards, which is a standard operational risk for this type of integration.
- Ingestion points: Telemetry metadata discovery tools such as
get-available-metrics,get-metric-labels, andget-trace-or-log-fields(SKILL.md). - Boundary markers: None explicitly specified for incoming metadata.
- Capability inventory: The skill utilizes the
create-dashboardMCP tool to commit changes to the platform (SKILL.md). - Sanitization: The instructions mandate the use of the
validate-dashboard-jsontool to verify the generated preset against the official schema prior to submission.
Audit Metadata