AWS Launches CloudWatch Omni, a Collaborative AI-Powered Observability Console for Engineering Teams
AWS introduced CloudWatch Omni, an off-console observability tool unifying application and AI-agent monitoring behind enterprise SSO, with an AI investigation agent.
Overview
Amazon Web Services introduced CloudWatch Omni on September 22, 2026, a new observability product that unifies monitoring of applications and AI agents in a single, off-console experience, according to AWS. Rather than requiring engineers to sign into the AWS Management Console, Omni is reached through a dedicated URL secured with enterprise single sign-on, and it layers an AI assistant, the Amazon DevOps Agent, into incident investigations alongside human engineers, as described in the AWS announcement.
What We Know
AWS built Omni on OpenTelemetry, so telemetry already flowing into CloudWatch “appears in Omni with nothing to reconfigure,” while other OpenTelemetry-instrumented workloads can send data to an OpenTelemetry Protocol endpoint, according to AWS. CloudWatch Omni is built on the same open standard that reached CNCF graduation earlier this year.
AWS frames the product around three problems it says engineering teams described to the company. First, collaboration: “Every engineer accesses CloudWatch Omni through a single URL with enterprise SSO (via IAM Identity Center, supporting Okta, Azure AD, and other providers). No AWS Console access is required,” AWS wrote, adding that SREs, developers, database engineers and managers work from the same investigation context so that when an incident escalates, the next person joining a session has full context already in front of them, according to AWS.
Second, adaptability: “CloudWatch Omni discovers your services, maps dependencies, and adjusts alarms automatically,” replacing manual dashboard curation and threshold tuning with topology that updates as teams deploy new services, according to AWS. Telemetry is organized by application rather than by infrastructure component, and each team gets a “Space” that groups the applications it owns along with associated telemetry, AWS said.
Third, AI-assisted investigation: “Amazon DevOps Agent participates alongside your team in investigation sessions, correlating signals and suggesting next steps,” working from the same telemetry engineers see so its suggestions are “grounded in the actual state of your application,” identifying correlated events across services and tracing root-cause paths through the dependency graph, according to AWS. Investigation history is captured automatically for post-incident review, AWS said.
Alongside the application-monitoring launch, AWS published a companion announcement describing Omni’s agent-observability capabilities, calling it “a purpose-built observability, evaluation, and experimentation solution for AI agents” meant to help teams “design, evaluate, and operate AI agents across any model provider, framework, or runtime,” according to AWS. AWS said agent behavior is non-deterministic, so “a prompt change can degrade response quality even when standard metrics show no errors,” and Omni responds by capturing every trace and running built-in evaluators for correctness, coherence, retrieval quality and tool selection, among others, the company said.
On the agent side, Omni ships through two surfaces: a native extension for VS Code and Kiro that surfaces traces as an agent runs, and a standalone web experience — separate from the AWS Management Console — for operators to monitor a fleet through SSO, according to AWS. The Kiro integration is one of two IDE extensions AWS lists as currently supported. An optional “Cloud Login” feature connects a local IDE environment to a customer’s AWS account to persist telemetry and share traces with a team; without it, Omni can run entirely locally during development, AWS said. The agent-observability surface includes 17 built-in evaluators for metrics such as coherence, helpfulness, faithfulness and routing correctness, a Trace Explorer with a “Compare mode” for viewing two traces side by side, a Session Explorer for multi-turn conversations, and an Agent Topology view that maps sub-agents, tools and their interconnections, according to AWS.
According to InfoWorld, Omni launches with support for agents built on LangGraph, CrewAI, OpenAI Agents SDK, Vercel AI SDK and AWS Strands, and it also supports independent evaluation tools including Braintrust, DeepEval and Ragas. InfoWorld reported that Omni is currently available in the US East (N. Virginia), US West (Oregon) and Europe (Ireland) regions, though AWS told the outlet that customers can centralize telemetry from other accounts and regions into a preferred Omni region “at no additional cost.” Pricing follows a metered structure: customers pay for data ingestion, storage and analytics, with ingestion charged per gigabyte and storage charged per gigabyte per month, and telemetry from AWS services billed at tiered pricing, according to InfoWorld.
Industry analysts quoted by InfoWorld were cautiously positive about the launch. Stephanie Walter, practice lead of AI Stack at HyperFrame Research, said, “With Omni, CIOs could gain one operating view across agents, applications, and infrastructure, in turn reducing tool fragmentation,” according to InfoWorld. Ashish Chaturvedi, executive research leader at HFS Research, told InfoWorld that “the blocker on enterprise agent deployment right now is rarely capability, because the agents can usually do the work,” and that “the blocker is that CIOs cannot confidently answer what happens when an agent gets it wrong, how they would know, and how fast they could find out” — a gap he said Omni could help close by giving teams “greater visibility into how agents behave when they make mistakes, how quickly those issues can be identified, and what impact they could have on business operations.”
What We Don’t Know
AWS has not published a specific per-gigabyte dollar figure for Omni’s ingestion or storage tiers in the material reviewed; both AWS and InfoWorld describe the pricing structure without listing exact rates. It is also not yet clear how quickly AWS plans to expand Omni beyond its three initial regions, or whether adoption among enterprises already committed to third-party observability platforms will be significant.
Analysis
The launch surfaces a tension that InfoWorld’s sources flagged directly. Chaturvedi cautioned that “the more of your observability runs through one vendor’s layer, the more your ability to understand your own systems depend on that vendor,” according to InfoWorld. Michael Leone, principal analyst at Moor Insights and Strategy, raised a cost concern specific to agentic workloads, telling InfoWorld that “agents generate a lot of telemetry because every prompt, tool call and handoff gets traced, so ingestion bills can climb faster than teams expect.” Leone also pointed to a gap in how enterprises define success for AI agents, saying “agent evaluations are only as good as an enterprise’s definition of a good answer, and a lot of them haven’t written that definition down yet,” per InfoWorld. Those caveats point to the same underlying bet AWS is making with Omni: that as agentic systems move from pilots into production, teams will accept concentrating both their application and agent telemetry inside a single vendor’s console in exchange for a unified view and less tool-switching during incidents.