Skip to main content

What's Supported in Harness AI SRE

Last updated on

This page outlines the supported platforms, integrations, and features available in Harness AI SRE.

For information about support in other Harness modules, go to the Harness Platform documentation.

Prerequisites​

To configure and use Harness AI SRE, ensure the following:

  • A Harness account with the AI SRE module enabled.
  • Administrator access to configure integrations, automations, and communication tools.
  • Service accounts with API access for third-party integrations (e.g., Jira, Slack, ServiceNow).

Integration capabilities​

Every AI SRE integration provides one of two capabilities, and some provide both:

  1. Ingest: The tool sends alerts, tickets, and change data into AI SRE.
  2. Automation: A runbook acts on the tool during a response.

Transport is how the connection is made. Webhooks are ingest-only. Connectors and delegates call the tool's API and power automation, on-call schedule sync, and polling-based ingestion. Go to Set up integration management to review the authoritative per-tool capability matrix.

Ingest integrations​

These tools send alerts, monitoring signals, and change data into AI SRE. Most use webhooks.

Monitoring and observability​

  • AlertSite
  • BigPanda
  • Datadog
  • Dynatrace
  • Grafana
  • Grafana Incident
  • New Relic
  • Opsgenie
  • Prometheus
  • Sentry

CI/CD and DevOps tools​

  • Bitbucket
  • GitHub
  • GitLab
  • Jenkins
  • Octopus Deploy
  • Travis CI

Incident management and ticketing tools​

  • Jira
  • PagerDuty
  • ServiceNow

Custom webhook sources​

  • Any tool capable of sending HTTP webhooks with JSON payloads can integrate with AI SRE.

Automation integrations​

These tools are acted on by runbooks through Harness connectors or delegates. The capability listed after each tool is what a runbook can do with it.

Communication and collaboration​

  • Slack: Create channels, invite users, manage user groups, execute runbooks via slug commands (/harness run <slug>)
  • Microsoft Teams: Create and select channels and teams
  • Google Chat: Link spaces to incidents, send messages via runbook actions
  • Zoom: Create meetings

ITSM and ticketing tools​

  • Jira: Create and manage Jira issues
  • ServiceNow: Create, update, and select records

On-call and escalation management​

  • Opsgenie: Acknowledge alerts, add responders to incidents
  • PagerDuty: Create incidents, add responders, manage on-call schedules
  • Service Paging Webhook: Trigger on-call notifications via HTTP POST or email from external monitoring tools (AI SRE native, no connector required)

Documentation and collaboration​

  • Confluence: Create, retrieve (as HTML), update, delete pages

DevOps and CI/CD automation​

  • GitHub: Create issues, comment on PRs, list commits
  • Harness: Execute pipelines, select input sets and projects

AI and machine learning capabilities​

Harness AI SRE leverages AI-driven incident response with:

  • AI-powered incident summarization: Generates real-time summaries from incident discussions.
  • RCA Change Agent: Automatically analyzes deployments, pull requests, ServiceNow change records, and change events to identify root cause candidates with confidence scores.
  • ServiceNow change correlation: When a ServiceNow connector exists, change request records automatically appear as root cause theories in active incidents.
  • AI-native post-mortem generation: Automatically generates structured post-incident reviews when incidents are closed, synthesizing incident metadata, timeline events, RCA theories, and notes into six standardized sections (Summary, Impact, Root Cause, Resolution, Insights, Lessons Learned).
  • Real-time action item detection: Automatically detects action items during active incidents from Slack conversations, meeting transcriptions, and incident notes, including assignee and due date information extracted from conversation context.
  • Investigator agent pipelines (Early Access): Extend the RCA Change Agent's investigation with custom investigation logic using pipeline stages to connect domain-specific data sources (internal wikis, observability tools, custom APIs) and enrich incident investigations with infrastructure-specific context.
note

Voice Transcription Analysis used in conjunction with Recall.ai

Stakeholder communication​

AI SRE keeps stakeholders informed during incidents through the following channels:

  • Status updates: Incident commanders can send structured status updates via email to stakeholders subscribed to impacted services. Recipients are resolved dynamically from service subscriber lists configured in the Service Directory.
  • Service subscriptions: Individual users and Harness User Groups can subscribe to services to receive automated status update notifications during incidents affecting those services.
  • Email delivery: Status updates are delivered via branded HTML emails from aisre-noreply@harness.io.

Feature flags​

Some Harness AI SRE features are released behind feature flags for early adopters.

To enable feature flags in your Harness account, contact Harness Support.

Next steps​

  • Go to Set up integration management to review the per-tool capability matrix and connect your tools.
  • Go to Runbooks to automate actions across the supported integrations.
  • Go to Incidents to see how ingested alerts become tracked incidents.
Tags:
Last updated on