AWS Quick and New Relic automate incident triage
AWS has released a template for building an incident triage assistant using Amazon Quick and New Relic's Model Context Protocol Server. The agent automates evidence gathering, root cause analysis, and task creation in a single conversational workflow. Internal testing showed the tool reduced the evidence-gathering phase of incident triage, lowering mean time to resolution and improving consistency across on-call rotations.
TL;DR
- Amazon Quick now integrates with New Relic's MCP Server to automate incident triage workflows
- The agent uses five New Relic reasoning tools to investigate incidents, quantify user impact, and surface error signatures
- A single prompt triggers investigation, RCA brief generation with evidence links, and Asana task creation for handoff
- New Relic's internal testing showed faster incident resolution and reduced knowledge loss between engineering shifts
Why It Matters
Incident triage is time-sensitive work that typically requires SREs and support engineers to manually collect evidence across separate tools. Automating this workflow through an agentic assistant reduces friction, standardizes investigation practices, and accelerates the critical early phase of incident response where speed directly impacts business continuity.
Business Impact
Reducing mean time to resolution (MTTR) directly improves business impact by minimizing service downtime and associated revenue loss. Automating evidence gathering also reduces the risk of knowledge loss between shifts and ensures consistent investigation standards, which lowers operational risk and improves team efficiency.
Key Implications
- AI agents can now orchestrate multi-tool incident response workflows, reducing manual handoffs and context switching for on-call engineers
- Native integrations between observability platforms and AI orchestration tools are becoming standard, enabling more sophisticated automation patterns
- Standardized incident triage processes through agentic workflows may improve organizational learning and reduce repeat incidents
What to Watch
Monitor adoption patterns among engineering teams using this template to understand whether agentic incident triage becomes a standard practice. Watch for extensions to this pattern, such as integration with additional observability platforms, ticketing systems, or automated remediation workflows. Track whether organizations report measurable improvements in MTTR and incident resolution consistency.
Subscribe to the newsletter
The latest stories and analysis, delivered to your inbox.
Free. No spam. Unsubscribe any time.
