grafana-mcp-tools
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external Grafana sources (metrics, logs, and dashboard metadata) which could contain attacker-controlled content.
- Ingestion points: Data entering the agent via tools like
query_loki_logs,query_prometheus, andsearch_dashboards(SKILL.md). - Boundary markers: No specific delimiters or instructions to ignore embedded commands are provided for tool outputs.
- Capability inventory: The skill includes write capabilities such as
update_dashboard,create_incident, andcreate_alert_rule(SKILL.md). - Sanitization: No explicit sanitization or validation of the external data is described.
- [SAFE]: The skill references the official Grafana GitHub repository and documentation for setup and resources.
- [SAFE]: Secret management is correctly handled by instructing users to use environment variables for sensitive tokens (e.g.,
GRAFANA_SERVICE_ACCOUNT_TOKEN) rather than hardcoding credentials. - [SAFE]: The skill explicitly instructs the agent not to attempt automated installation of the MCP server, ensuring a human-in-the-loop process.
Audit Metadata