Grafana for AI agents
Grafana is an open and composable observability platform for visualizing metrics, logs, and traces on dashboards, widely used by engineering teams to monitor infrastructure and application health. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Grafana.
Check platform health status
Check Grafana's own health endpoint to confirm the instance is up before debugging a dashboard.
Query a public dashboard panel
Query a shared dashboard panel's data to pull a metric into a status report.
Check license status
Look up the current license status, read-only, before renewing or upgrading a plan.
11 Grafana actions, graded by risk.
Every Grafana action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
9Look things up. Allowed by default.
- grafana.get_healthGet Grafana Health Status
- grafana.get_statusGet License Status
- grafana.get_ruler_ringGet Ruler Ring Status
- grafana.get_index_gateway_ringGet Index Gateway Ring Status
- grafana.query_public_dashboardQuery Public Dashboard Panel
- grafana.get_store_gateway_tenantsGet Store Gateway Tenants
- grafana.get_distributor_ha_trackerGet Distributor HA Tracker
- grafana.get_overrides_exporter_ringGet overrides exporter ring
- grafana.retrieve_jwksRetrieve JWKS
write
2Create and change things. Allow, or ask the user first.
- grafana.post_acsPost ACS
- grafana.create_otlp_v1_logsCreate OTLP v1 Logs
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
Grafana in three steps.
- 01Your users connect GrafanaThey add their Grafana bearer token on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “create OTLP v1 Logs” can wait for the user to approve.
- 03Any agent can actYour agent calls Grafana through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('grafana', { read: 'allow', write: 'ask', // create_otlp_v1_logs destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Grafana connects.
Users add their Grafana bearer token on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Grafana connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- Bearer token
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Grafana from any agent.
Grafana and Arc0, answered.
Can I use Grafana with Claude, ChatGPT or Cursor?
Yes. Connect Grafana to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Grafana actions you allow.
How do users connect Grafana?
Users add their Grafana bearer token on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Grafana actions can my agent take?
11 in total: 9 read, 2 write and 0 destructive, such as “create OTLP v1 Logs”. Your policies decide which of them each agent may call.
Can I make my agent read-only in Grafana?
Yes. Allow read actions and deny writes in the Grafana policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call Grafana too?
Yes. The same Grafana connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Grafana into your agent.
Your users connect Grafana once, under your brand. Your agent gets 11 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan