Grafana MCP integration for AI agents.
Connect AI agents to Grafana through 19 structured actions, including get alert rule, create dashboard, and delete dashboard. Review authentication, inputs…
- ACTION CONTRACTS
- 19
- AUTHENTICATION
- API key
- SOURCE STATUS
- Provider-backed
CATALOG SOURCE REVIEWED AUGUST 23, 2026 / ACTION NAMES AND SCHEMAS DERIVED FROM WORKING MACHINES PROVIDER SOURCE
CAPABILITY PROFILE
What the Grafana integration exposes
The Grafana Agent App exposes 19 provider-backed actions for data and developer tools work. Its current contract lets an authorized agent retrieve one grafana-managed alert rule by uid via the provisioning api, create a grafana dashboard resource in a namespace, and delete a grafana dashboard resource by uid. These operations are called through Working Machines as typed capabilities rather than through browser navigation or copied UI steps.
A connection uses API key. Before execution, the agent can inspect the selected action, its required fields, declared scopes, and expected output contract. Provider credentials remain inside the Working Machines runtime; the calling agent receives the capability and its structured result, not the underlying secret.
For reliable operation, start with the narrowest action that satisfies the task, resolve stable provider identifiers before changing state, and validate the returned object or status after execution. Availability still depends on the connected Grafana account, granted provider permissions, workspace policy, region, plan, and upstream API behavior.
VERIFIED ACTION SAMPLE
Real Grafana capabilities.
Showing 12 of 19 actions. Risk labels are conservative signals based on operation names, not substitutes for provider documentation or runtime policy.
get_alert_ruleREADRetrieve one Grafana-managed alert rule by UID via the provisioning API.
INPUTS: uid
create_dashboardWRITECreate a Grafana dashboard resource in a namespace.
INPUTS: uid / generateName / folderUid / spec / namespace
delete_dashboardHIGH IMPACTDelete a Grafana dashboard resource by UID.
INPUTS: uid / namespace
create_data_sourceWRITECreate a Grafana data source using a JSON payload accepted by Grafana.
INPUTS: dataSource
create_folderWRITECreate a Grafana folder in a namespace.
INPUTS: title / uid / generateName / parentUid / namespace
delete_data_sourceHIGH IMPACTDelete a Grafana data source by UID.
INPUTS: uid
delete_folderHIGH IMPACTDelete a Grafana folder by UID.
INPUTS: uid / namespace
get_dashboardREADRetrieve one Grafana dashboard resource by UID.
INPUTS: uid / namespace
get_data_sourceREADRetrieve one Grafana data source by UID.
INPUTS: uid
get_folderREADRetrieve one Grafana folder by UID.
INPUTS: uid / namespace
list_alert_instancesREADList currently firing or pending Grafana alert instances from the built-in Alertmanager.
INPUTS: active / silenced / inhibited
list_alert_rulesREADList all Grafana-managed alert rules via the provisioning API.
INPUT CONTRACTS
Know what the action needs before it runs.
get_alert_rule
uidREQUIRED- The Grafana alert rule UID.
create_dashboard
uidOPTIONAL- Optional explicit Grafana dashboard UID.
generateNameOPTIONAL- Optional UID prefix Grafana can use to generate a dashboard UID.
folderUidOPTIONAL- Optional folder UID for the new dashboard.
specREQUIRED- The Grafana dashboard spec JSON. This is forwarded to Grafana as the dashboard body.
namespaceOPTIONAL- The Grafana API namespace. Use default for the main organization.
delete_dashboard
uidREQUIRED- The Grafana dashboard UID.
namespaceOPTIONAL- The Grafana API namespace. Use default for the main organization.
create_data_source
dataSourceOPTIONAL- The Grafana data source payload. Use official Grafana data source fields such as name, type, access, url, jsonData, and secureJsonData.
PROVIDER-SPECIFIC WORKFLOWS
Jobs this Agent App can support
Inspect Get Alert Rule
Retrieve one Grafana-managed alert rule by UID via the provisioning API. Use this as a bounded discovery step, retain the returned identifier, and avoid expanding the read beyond the task's stated scope.
get_alert_ruleControl Create Dashboard
Create a Grafana dashboard resource in a namespace. Resolve the target first, present material changes for confirmation, and make retries idempotent where the provider supports it.
create_dashboardVerify with Get Dashboard
Retrieve one Grafana dashboard resource by UID. Compare the returned provider state with the intended outcome and preserve stable IDs or canonical links in the run record.
get_dashboardSAFETY BOUNDARY
Operate Grafana with explicit limits
- Authorize Grafana with API key and grant only the provider access required by the selected actions.
- 9 actions are change-capable by name. Confirm the target identity and material parameters before allowing a write.
- Treat `delete_dashboard`, `delete_data_source`, and `delete_folder` as high-impact operations and require an explicit approval boundary.
- No provider scope string is declared on the sampled actions. Verify the connected account's actual permissions in Grafana rather than assuming unrestricted access.
- After a call, inspect the structured result and execution record before reporting that the Grafana task completed successfully.
CONNECTION MODEL
API key
Grafana service account token sent as a Bearer token. Create one in Administration > Users and access > Service Accounts: https://grafana.com/docs/grafana/latest/administration/service-accounts/
Working Machines stores provider credentials behind the execution boundary. An agent can use an authorized connection identity, but catalog discovery alone does not reveal OAuth tokens, API keys, or provider secrets.
- READ SIGNALS
- 10
- WRITE SIGNALS
- 6
- HIGH IMPACT
- 3
- REVIEW SIGNALS
- 0
EVIDENCE AND AVAILABILITY
Provider reference
Action names, input fields, authentication types, and counts on this page are generated from the Working Machines provider catalog. Provider behavior, quotas, object semantics, account eligibility, and regional availability remain governed by Grafana.