loki
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a reference for Grafana Loki and LogQL. All code snippets are illustrative and use placeholders for sensitive information (e.g., "user:apikey").
- [SAFE]: External references (e.g., "grafana.com") point to official, well-known service documentation.
- [SAFE]: The skill promotes secure configuration by recommending the use of environment variables for authentication and redacting sensitive data from logs.
- [INDIRECT_PROMPT_INJECTION]: The skill describes log parsing techniques (JSON, Logfmt, Regexp) which process untrusted external data. While this creates a potential attack surface if an agent analyzes malicious log content, the skill itself is purely instructional and explicitly mentions redaction as a security measure.
- Ingestion points: Log files ("/var/log/app/*.log") and Kubernetes pod logs defined in "SKILL.md".
- Boundary markers: Not applicable for a documentation-only skill.
- Capability inventory: No tool execution or code execution capabilities are defined or requested by the skill.
- Sanitization: "ASSESSMENT.md" explicitly requires that sensitive data is redacted before storage or query exposure.
Audit Metadata