Documentation Grafana Cloud Platform Workflows Reference Blueprints

Blueprints

Blueprints are prebuilt workflow definitions for common automation, such as kicking off a major incident, welcoming the incoming on-call, or posting a weekly report. Each Grafana Workflows blueprint is a complete workflow that you finish configuring for your team. Use the pages in this section to see what a blueprint does, what it needs, and its full YAML definition.

How to use blueprints

You can start from a blueprint in the Workflows app or from its YAML definition.

In the Workflows app, click Create workflow and choose a blueprint from the library. The editor opens with the blueprint’s steps filled in. For details, refer to Start from a blueprint.

To manage workflows as code, copy the YAML definition from a blueprint’s page and use it with gcx, the Workflows API, or Terraform. Coding agents can also use the definitions as examples of how workflows are built. For details, refer to Manage workflows as code.

A workflow created from a blueprint starts disabled. Customize the workflow, then turn it on.

What blueprints need from you

Blueprints are starting points, not finished workflows. Most need some information or setup from you before they can run:

• Values for your team: Many steps leave inputs empty, such as a Slack channel ID, an IRM schedule name, or a URL. Each step’s description says what to set.
• Secrets: Blueprints that call an external API, such as Jira, GitHub, or the Grafana HTTP API, reference a workflow secret that holds an API token. Create the secret before you run the workflow. For details, refer to Manage secrets.
• Integrations: Blueprints that post to Slack or Microsoft Teams need that integration connected to Grafana IRM. Blueprints that react to incidents, alert groups, or on-call schedules need Grafana IRM.
• Grafana Assistant: Some blueprints use Grafana Assistant to investigate alerts, summarize incidents, or write reports, so they need Grafana Assistant in use on your stack.

For the exact requirements of a specific blueprint, refer to its page in this section.

Browse blueprints

Find a blueprint for the automation you want to set up.

• AI alert triage
  Grafana Assistant investigates every new alert group, posts a triage summary to Slack, and declares an incident when it decides the alert needs escalating.
• Annotate dashboards with incidents
  When an incident is resolved, add an annotation that spans the incident to your Grafana dashboards, so it lines up with your metrics.
• Announce on-call changes in Microsoft Teams
  Announce on-call changes in a Microsoft Teams channel so everyone knows who to contact.
• Archive incident channel after resolve
  Archive an incident's Slack channel a day after it's resolved, with a heads-up first.
• Assign commander from schedule
  When an incident is declared without a commander, assign whoever is on call in your incident commander schedule.
• Broadcast key updates to Microsoft Teams
  Keep stakeholders informed by sharing every key update on a critical or major incident in a Microsoft Teams channel.
• Create Jira issue for incident follow-ups
  When a critical or major incident is resolved, open a Jira issue to track follow-up work and link it on the incident timeline.
• Declare incident for critical alerts
  Declare a Grafana IRM incident automatically when a new alert group carries a severity=critical label.
• Draft post-incident review
  When a critical or major incident is resolved, Grafana Assistant drafts a post-incident review as a Grafana Notebook, then shares an executive summary and the notebook link on the incident timeline and in Slack.
• Enforce severity before resolve
  Stop incidents from being resolved while their severity is still Pending. Reopens the incident and tells the team why.
• Escalate severity changes
  React to severity changes. Page leadership and alert the exec channel when an incident becomes Critical, and give a heads-up when one escalates to Major.
• Invite on-call from alert group label
  When someone adds a team label to an alert group, invite that team's current on-call engineer to the alert group.
• Invite on-call from incident label
  When someone adds a team label to an active incident, invite that team's current on-call engineer to the incident.
• Learn: expressions and branching
  A runnable tour of CEL expressions, step outputs, switch branches, and filters. No setup or integrations needed. Click Run and try different payloads.
• Learn: shape data with Python
  A runnable example of the Python step. Reshapes a list of alerts, finds noisy services, and passes the result to the next step as JSON. No setup needed.
• Major incident kickoff
  Spin up a major incident in seconds. Creates the incident Slack channel, assigns the on-call incident commander, posts a help message, and bookmarks the runbook.
• Microsoft Teams incident bridge
  Open a Microsoft Teams bridge for every critical incident and announce it in your Teams incident channel with a join button.
• Morning operations briefing
  Every weekday morning, Grafana Assistant posts a short Slack briefing on active incidents, firing alerts, and what happened overnight.
• On-call handoff report
  When the on-call rotation changes, Grafana Assistant writes a handoff report from the past week of incidents and alert groups and posts it to Slack.
• On-call notification audit
  Every Monday, check that everyone has a verified phone number and that their important notification rules can actually reach them, then post the gaps to Slack.
• Require incident fields on resolve
  When an incident is resolved, check for required custom fields and labels, and nudge the team to fill in anything that is missing.
• Route alerts to team channels
  Send each new alert group to the Slack channel of the team that owns it, based on its team label, with a fallback channel for everything else.
• Run GitHub Actions remediation
  Run a GitHub Actions remediation workflow, such as a service restart, when an alert group labeled for automatic remediation fires.
• Vendor status page watch
  Watch a vendor's public status page and post to Slack when they open a new incident, so you hear about upstream outages before your users do.
• Weekly incident metrics to Microsoft Teams
  Every Monday, pull last week's incidents from the Grafana IRM API, compute counts and time to resolve in Python, and post a scorecard to Microsoft Teams.
• Weekly service health summary
  Every Friday morning, Grafana Assistant reviews the health of the services you own and posts a summary to Slack.
• Welcome the incoming on-call
  Send a Slack direct message to whoever just went on call, with their shift end time and a start-of-shift checklist.