skills/grafana/skills/grafana-oss/Gen Agent Trust Hub

grafana-oss

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides legitimate documentation and configuration templates for Grafana OSS core features such as dashboards, alerting, and data sources.
  • [CREDENTIALS_UNSAFE]: Security best practices are followed by using generic placeholders (e.g., 'secret', 'yourpassword') or environment variable references (e.g., '$GRAFANA_MYSQL_PASSWORD') in all configuration examples and API calls.
  • [COMMAND_EXECUTION]: Includes standard examples for administrative tasks using 'curl' for the Grafana API and 'grafana-cli' for plugin installation.
  • [EXTERNAL_DOWNLOADS]: Network references are limited to trusted and well-known services including Grafana's official documentation, Okta for authentication, and standard SMTP/Slack endpoints.
  • [DATA_EXFILTRATION]: No patterns of unauthorized data harvesting or network exfiltration were identified.
  • [SAFE]: Regarding Indirect Prompt Injection (Category 8): 1. Ingestion points: Data source query results from systems like Prometheus and Loki are processed for visualization in panels and alerts. 2. Boundary markers: Absent in provided configuration snippets. 3. Capability inventory: Administrative actions via Grafana API (curl) and plugin management (grafana-cli). 4. Sanitization: Performed by Grafana's internal data handling and query sanitization mechanisms. This represents standard and expected observability platform behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 04:38 PM