Curated Automation Remediation: From EDA to Orchestrated Automation

AIOps Use Case

Pattern guide, not a partner integration.

Observability and ITSM partners in examples (Splunk, Instana, ServiceNow) show integration points; the pattern is the approved automation library, not a single vendor.

Overview

Teams need consistent remediation, but unconstrained AI agents may generate a different fix every time the same incident recurs. Curated automation remediation keeps execution in a library of approved workflows while AI selects the right playbook from that library under guardrails and approvals.

Challenge: Speed of AI suggestions without sacrificing trust, auditability, or change control.

Solved with AI + automation: Automation teams maintain approved remediation workflows. AI intercepts incidents and chooses from curated options. AAP executes through governed, auditable processes.

Business outcomes: Consistent execution, faster remediation, improved auditability, reduced operational risk.

Buyer question:

How do we remediate faster using automation teams already trust?

Background

This use case sits at Walk maturity: event- or ticket-initiated flows that combine EDA, inference, and job templates. It pairs naturally with Red Hat Lightspeed and pre-approved content at Crawl/Walk and with Automation code assistant at Run only inside policy boundaries – selection from curated content remains the default for production.

For a shared high-level Detect → MCP → select → execute walkthrough across partners, see AIOps automation with Ansible. This page is the dedicated pattern guide and will expand with deeper scenarios, demos, and Arcade content.

Solution

EDA triggers on observability or ITSM events (or an operator starts from an AI client)
AAP MCP exposes the approved automation library – job templates and workflows the AI can search like a menu
AAP runs only approved job templates and workflows selected through MCP (RBAC follows the operator token)
AI maps incident context to library entries (patch, restart, expand storage, rotate certificate, update firewall rule, and similar) without generating new playbooks at incident time
AO adds approval nodes and routing when multiple curated options match

Who Benefits

Persona	Challenge	What They Gain
Automation lead	Shadow fixes outside AAP	All remediation through the library
SRE	Repeated incidents, ad hoc scripts	Same fix every time, faster
Security / compliance	Unaudited AI-generated changes	Selection only; execution is governed

Prerequisites

Ansible Automation Platform 2.6.4 or later with MCP gateway access for job template discovery and launch (DevTools MCP setup)
Event-Driven Ansible for event-initiated selection flows
Automation orchestrator when approvals and branching are required

EDA to AO adoption path

Coming soon:

Walkthrough scenarios (EDA trigger, AI selection from library, approval, AAP job) will be added in a future revision.

Stage 1: EDA + AAP only

Static mapping from alert type to one approved job template.

Stage 2: AI enrichment

AI chooses among labeled job templates based on incident context.

Stage 3+: Automation orchestrator

Approval gates, switch nodes for confidence, and validation loops after remediation.

Validation

Symptom	Status
End-to-end scenario tests	Content in progress