kubesense-mcp
Pass
Audited by Gen Agent Trust Hub on Apr 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides comprehensive documentation and examples for interacting with KubeSense MCP observability tools for logs, traces, and metrics. All described functionality is limited to querying and aggregating cluster telemetry data.
- [SAFE]: No remote code execution, obfuscation, or data exfiltration patterns were identified. The tools use standard SQL-like filtering and PromQL, and do not execute arbitrary system commands.
- [PROMPT_INJECTION]: The skill defines a surface for processing external observability data (logs and traces), which could potentially contain adversarial instructions (Indirect Prompt Injection). However, given that the skill's capabilities are restricted to read-only querying and data aggregation, the risk is minimal.
- Ingestion points:
search-logs,search-traces,analyze-logs,analyze-traces(SKILL.md, kubesense-logs/SKILL.md, kubesense-apm/SKILL.md). - Boundary markers: The skill does not explicitly instruct the agent to use delimiters when processing retrieved log or trace content.
- Capability inventory: Limited to KubeSense MCP query tools; no file system writes, shell access, or administrative capabilities are exposed.
- Sanitization: No specific sanitization or filtering of log/trace content is mentioned before agent processing.
Audit Metadata