Grafana Loki Log Correlation Agent
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs legitimate monitoring and log correlation tasks using Grafana Loki and Tempo APIs. No malicious code, obfuscation, or unauthorized access patterns were detected in the instructions or installation commands.
- [PROMPT_INJECTION]: An indirect prompt injection surface was evaluated due to the processing of external log data:
- Ingestion points: The agent queries log lines from the Grafana Loki LogQL API.
- Boundary markers: None are explicitly defined in the skill instructions to delimit log content from instructions.
- Capability inventory: The skill performs network operations to monitoring APIs and generates structured JSON/YAML files (alerts and dashboards).
- Sanitization: No content sanitization or instruction-ignoring wrappers are specified for the processed logs.
- Note: This represents a standard risk for log processing tools and does not indicate malicious intent.
Audit Metadata