Microsoft Azure NetworkExperiments

★ Only Publicly Available OpenAPI DocumentDeveloper ToolsMonitoring Observabilityoauth214 Endpoints

Install Jentic One Beta

Connect the NetworkExperiments to your agent

Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the NetworkExperiments, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.

Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.

Step 1: Jentic One Host machine

# On the machine that will host your Jentic One instance:
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh

Step 2: Agent machine

# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh
jentic register       # connects your agent to your Jentic One instance

Jentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.

Capabilities

What an agent can do with NetworkExperiments API.

Create a NetworkExperiment Profile to host a set of related A/B experiments

Configure A/B experiments comparing candidate endpoints under a profile

Pull latency scorecards aggregated by region for any experiment

GET STARTED

Start building with NetworkExperiments API

Explore with Jentic

View OpenAPI Document

For Agents

Run real-user A/B latency experiments between candidate endpoints with Azure Internet Analyzer. Covers 14 endpoints across NetworkExperiment Profiles, Experiments, latency scorecards, time series, and reports.

Use for: Create a NetworkExperiment Profile in westeurope for my CDN evaluation, Run an A/B experiment comparing two CDN endpoints over 7 days, Get the latency scorecard for experiment cdn-eval grouped by country, Retrieve a 30-day time series of latency values for experiment cdn-eval

Not supported: Does not handle synthetic probing, application performance monitoring, or routing decisions — use for real-user A/B latency experiments and their reports only.

Jentic publishes the only available OpenAPI specification for Azure Network Experiments, keeping it validated and agent-ready. The Microsoft.Network NetworkExperiments resource provider, also known as Azure Internet Analyzer, lets engineering teams measure the real-world latency that end users see between their clients and a set of candidate endpoints. Operators define a NetworkExperiment Profile, configure one or more A/B experiments comparing endpoints, and pull back latency scorecards and time series data from real client measurements. The 2019-11-01 surface covers experiment lifecycle, preconfigured endpoints, latency scorecards, time series, and report generation.

Use Cases

Patterns agents use NetworkExperiments API for, with concrete tasks.

★ CDN selection through real-user latency

Performance engineering teams choosing between CDN providers run an Azure Internet Analyzer experiment that compares two or more CDN endpoints serving the same asset. Real client measurements aggregated into latency scorecards reveal which CDN delivers lower latency for the actual user population, by country and by ASN. A two-week experiment typically gathers enough samples to drive a confident decision in markets that matter most to the business.

Create profile cdn-eval in westeurope, then create experiment cdn-comparison comparing endpoints cdn-a.contoso.net and cdn-b.contoso.net, and after 14 days fetch the latency scorecard.

Origin migration validation

Teams migrating an origin between regions or providers want to validate that real-user latency does not regress before flipping production traffic. Running an experiment comparing the current origin to the candidate origin gives a quantitative answer rather than a synthetic-only signal. The latency time series surface trends over the migration window so operators can see whether performance is stable, improving, or regressing.

Create experiment origin-migration comparing origin-old.contoso.net and origin-new.contoso.net for two weeks, then retrieve the time series for daily review.

Periodic latency reporting

Performance teams produce a recurring report of network experiment outcomes for stakeholders. The reports endpoint generates cross-experiment summaries that can be exported into BI tooling. Combined with scheduled time series pulls, this produces a stable cadence of performance evidence to inform infrastructure decisions and SLA communications.

Generate the cross-experiment report for profile cdn-eval and export the latency scorecard to a CSV grouped by country code.

Agent-driven latency experiments

AI agents can use Jentic to discover NetworkExperiments operations by intent, load the structured input schema, and execute calls against management.azure.com without browsing the Azure REST docs. A performance agent can launch an experiment comparing two candidate endpoints from a single intent, monitor the latency scorecard each day, and surface a recommendation when the data is statistically meaningful.

Search Jentic for create azure network experiment, load the input schema, and execute it for endpoints a.example.com and b.example.com under profile cdn-eval in westeurope.

Key Endpoints

14 endpoints — jentic publishes the only available openapi specification for azure network experiments, keeping it validated and agent-ready.

METHOD

PATH

DESCRIPTION

PUT

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}

Create or update a NetworkExperiment Profile

PUT

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}/Experiments/{experimentName}

Create or update an A/B experiment

GET

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}/Experiments/{experimentName}/LatencyScorecard

Get the latency scorecard for an experiment

GET

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}/Experiments/{experimentName}/Timeseries

Get the latency time series for an experiment

GET

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}/PreconfiguredEndpoints

List preconfigured endpoints for use in experiments

PUT

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}

Create or update a NetworkExperiment Profile

PUT

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}/Experiments/{experimentName}

Create or update an A/B experiment

GET

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}/Experiments/{experimentName}/LatencyScorecard

Get the latency scorecard for an experiment

GET

/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}/Experiments/{experimentName}/Timeseries

Get the latency time series for an experiment

Why Jentic?

Three things that make agents converge on Jentic-routed access.

Credential isolation

Your Azure AD credential is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.

Intent-based discovery

Agents search Jentic by intent such as 'create a network experiment' or 'get the latency scorecard', and Jentic returns the matching ARM operation with its input schema so the agent calls the correct path without browsing Microsoft Learn docs.

Time to first call

FAQs

Specific to using NetworkExperiments API through Jentic.

Why is there no official OpenAPI spec for Azure Network Experiments?

Microsoft Azure does not publish an OpenAPI specification for the NetworkExperiments slice of Microsoft.Network. Jentic generates and maintains this spec from the Azure Resource Manager contracts so that AI agents and developers can call Azure Internet Analyzer via structured tooling. It is validated against the live API and kept up to date. Get started at https://app.jentic.com/sign-up.

What authentication does the Azure Network Experiments API use?

The API uses Azure Active Directory OAuth 2.0 with the user_impersonation scope against https://login.microsoftonline.com/common/oauth2/authorize. Jentic stores the AAD client secret or service principal credentials encrypted in its vault and supplies scoped bearer tokens to agents at execution time without exposing the underlying credential.

Can I run an A/B latency experiment between two endpoints through this API?

Yes. After creating a NetworkExperiment Profile, PUT under /providers/Microsoft.Network/NetworkExperimentProfiles/{profileName}/Experiments/{experimentName} creates an A/B experiment that defines control and treatment endpoints. Real client measurements accumulate over the experiment duration and are exposed via the LatencyScorecard and Timeseries subresources.

What are the rate limits for the Azure Network Experiments API?

The control plane follows the standard Azure Resource Manager throttling limits, which apply per subscription and per region. Read operations are limited to 1200 requests per hour per subscription. Latency scorecard and time series reads are inexpensive but should be cached for at least one hour, since the underlying real-user data updates on a slow cadence.

How do I retrieve a latency scorecard through Jentic?

Search Jentic for get azure network experiment latency scorecard, load the schema for GET .../Experiments/{experimentName}/LatencyScorecard, and execute it with the profile and experiment names. Jentic handles AAD token exchange and returns the scorecard JSON, which the agent can then summarise for stakeholders.

Does the API run synthetic probes or use real user measurements?

Azure Internet Analyzer uses real-user measurements collected through a JavaScript snippet in customer web pages, not synthetic probes. The experiment results therefore reflect the latency seen by the actual visitor population, which is the property that drives the API's value over synthetic monitoring.