AI agents call search_dashboards to retrieve information from Grafana without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a read-only query against Grafana dashboards, returning metadata (uid, title, folder). It has no capability to modify, delete, or execute actions. The broader server context confirms 'read-only access' design. Severity is low because dashboard discovery itself poses minimal risk; the actual dashboard content retrieval requires a separate get_dashboard call.
From the tool's definition Tool description states 'Search Grafana dashboards by query, tags, or type. Returns uid/title/folder' — a search operation that retrieves dashboard metadata without modification or side effects.
Attacks that exploit this kind of access
Search Grafana dashboards by query, tags, or type. Returns uid/title/folder — use get_dashboard with uid for details. It is categorised as a Read tool in the Grafana MCP Server, which means it retrieves data without modifying state.
Register the Grafana MCP server in PolicyLayer and add a rule for search_dashboards: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Grafana. Nothing to install.
search_dashboards is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the search_dashboards rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for search_dashboards. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
search_dashboards is provided by the Grafana MCP server (ninanung/grafana-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →