Grafana

You are working with an MCP server that enables AI-powered, programmatic integration with Grafana. It exposes Grafana’s dashboards, data sources, metrics, logs, alerting, and administrative functions through standardized MCP tools so you can automate and orchestrate Grafana tasks from your MCP client.

How to use

You will interact with the Grafana MCP Server through an MCP client. The server provides a rich set of tools to manage dashboards, query Prometheus metrics, search Loki logs, handle alerts, manage users and organizations, and generate deep links to Grafana resources. Use these tools to perform common tasks such as finding dashboards, running PromQL queries, testing datasource connections, deploying and evaluating RAMP sensor data, and generating URLs for dashboards and explore views. Your MCP client can call the tools directly, pass structured arguments, and receive typed responses that you can act upon in your automation workflows.

How to install

Prerequisites you need before installation:

• Node.js 18+ with ES modules support

• TypeScript 4.9+

• Grafana v8.0+

Step-by-step setup workflow you can follow:

# Clone the repository
git clone https://github.com/your-org/grafana-mcp
cd grafana-mcp

# Install dependencies
npm install

# Build the project
npm run build

Configuration

Configure the server using environment variables. Create a .env file or export these variables in your environment.

# Required Configuration
GRAFANA_URL=https://your-grafana-instance.com
GRAFANA_TOKEN=your-service-account-token-or-api-key

# Optional Configuration
GRAFANA_DEBUG=false                    # Enable debug logging
GRAFANA_TIMEOUT=30000                  # HTTP timeout in milliseconds
GRAFANA_DISABLE_TOOLS=admin,alerting   # Disable specific tool categories

# TLS Configuration (Optional)
GRAFANA_TLS_CERT_FILE=/path/to/client.crt
GRAFANA_TLS_KEY_FILE=/path/to/client.key
GRAFANA_TLS_CA_FILE=/path/to/ca.crt
GRAFANA_TLS_SKIP_VERIFY=false          # Skip certificate verification (insecure)

Security

Security is designed to protect credentials and minimize risks. The server sanitizes data in logs and errors, categorizes errors for safe user-facing messages, enforces TLS with certificate validation, and operates tools with the least privileges necessary.

Examples

Here are practical examples you can adapt to your environment. These illustrate common usage patterns without exposing protocol details.

# Example: start the server in development mode with hot reload
npm run dev

# Example: run tests
npm test

# Example: lint and format
npm run lint
npm run format

Notes

The configuration and startup steps assume you have a Grafana instance accessible at the URL you provide and a valid token. Adjust timeouts and TLS settings to match your security requirements and network conditions.

Troubleshooting

If you encounter connectivity or authentication issues, verify the Grafana URL and token, check network access, and review logs for sanitized error messages that indicate credential or permission problems.

Development

Development supports hot-reload during development, test execution, and code quality checks. Ensure you run the appropriate npm scripts and validate TypeScript types during integration.

Available tools

search_dashboards

Find dashboards by title, tags, or metadata

get_dashboard_by_uid

Retrieve complete dashboard details

get_dashboard_panel_queries

Extract queries from all panels

update_dashboard

Create or update dashboards

get_dashboard_versions

View dashboard version history

restore_dashboard_version

Restore to a specific version

delete_dashboard

Remove dashboards

list_datasources

List all configured datasources

get_datasource_by_uid

Get datasource details by UID

get_datasource_by_name

Get datasource details by name

test_datasource_connection

Test datasource connectivity

get_datasources_by_type

Filter datasources by type

get_default_datasource

Get the default datasource

check_datasource_exists

Check if a datasource exists

query_prometheus

Execute PromQL queries

get_prometheus_metadata

List available metrics

get_prometheus_labels

Get label names

get_prometheus_label_values

Get values for specific labels

get_prometheus_series

Find time series

build_prometheus_query

Interactive query builder

query_loki

Execute LogQL queries

get_loki_labels

Get log stream labels

get_loki_label_values

Get label values

get_loki_series

Find log series

build_logql_query

LogQL query builder

get_loki_stats

Ingestion statistics

list_alert_rules

List all alert rules

get_alert_rule

Get specific alert rule details

create_alert_rule

Create new alert rules

update_alert_rule

Update existing rules

delete_alert_rule

Delete alert rules

list_contact_points

List notification contact points

get_contact_point

Get contact point details

test_contact_point

Test notification delivery

list_alert_rule_groups

List alert rule groups

list_teams

List teams

get_team_by_uid

Get team details

list_users

List organization users

get_current_user

Get current user info

list_folders

List dashboard folders

get_folder_by_uid

Get folder details

list_api_keys

List API keys

list_service_accounts

List service accounts

get_current_organization

Get current organization info

generate_deeplink

Generate deeplink URLs for dashboards, panels, or Explore

generate_dashboard_url

Generate dashboard URLs with time range and variables

generate_panel_url

Generate panel-specific URLs

generate_explore_url

Generate Explore view URLs

generate_prometheus_explore_url

Generate Prometheus Explore URLs

generate_loki_explore_url

Generate Loki Explore URLs

get_time_range_presets

Get common time range presets

validate_time_range

Validate time range parameters

discover_sensors

Auto-discover Corelight sensors via SSH tunnels (ports 8080-8099)

sensor_status

Get health and version info for a sensor

query_sensor_metric

Query Prometheus metrics from a sensor's Grafana

deploy_ramp_dashboard

Deploy a standard RAMP dashboard to a sensor

list_baselines

List available performance baselines for comparison

sensor_performance_verdict

Compare current metrics against baselines (pass/marginal/fail)

annotate_test

Create Grafana annotations to mark test events