AWS Security ChangesHomeSearch

AWS AmazonCloudWatch: Clarify local dev credential source in CloudWatch Omni agent guide

Service: AmazonCloudWatch · 2026-11-20 · Documentation medium

File: AmazonCloudWatch/latest/monitoring/omni-agents-close-the-loop.md · Type: iam

Summary

Reworked the agent-development walkthrough: step 4 now states local development uses the credentials from the Local dev environment card even after space sign-in; step 9 changes the guidance from leaving 'Use this account for local dev' selected to clearing it, so the playground and evaluators run locally with the AWS account connected in the Local dev environment card rather than the space's sign-in account. Remaining edits are cosmetic UI renames ('Agent Dashboards' -> 'Agent overview', status bar -> 'Data source' control) and identifier changes (scope-regression -> scope_regression, On-Topic Compliance -> OnTopicCompliance).

Security assessment

The changed line reverses prior guidance and explicitly scopes which AWS credentials are used for local agent execution (the account in the Local dev environment card, not the space sign-in account), which affects credential handling and account isolation during development. No vulnerability, CVE, or incident is referenced, so it is security-adjacent documentation rather than a fix.

Evidence

+  * Clear **Use this account for local dev**. The playground and evaluators in steps 11 and 12 run on your machine with the AWS account connected in the **Local dev environment** card, not with your space sign-in. If you did not connect one in step 4, choose **Connect** in that card, choose **Automatic** for **Method** , choose your Region, and then choose **Connect**.

Diff

diff --git a/AmazonCloudWatch/latest/monitoring/omni-agents-close-the-loop.md b/AmazonCloudWatch/latest/monitoring/omni-agents-close-the-loop.md
index 3ef6b84..e41b604 100644
--- a//AmazonCloudWatch/latest/monitoring/omni-agents-close-the-loop.md
+++ b//AmazonCloudWatch/latest/monitoring/omni-agents-close-the-loop.md
@@ -104 +104 @@ Steps 1–3 happen in your editor.
-  4. Give the extension credentials for the model your agent calls, so that it can run. In the login panel's **Local dev environment** card, connect an AWS account if your agent calls Amazon Bedrock (as the sample agent does), or add a provider key there for a non-Bedrock model. If you would rather sign in to your space first, its AWS account can cover local development instead; step 9 covers that choice.
+  4. Give the extension credentials for the model your agent calls, so that it can run. In the login panel's **Local dev environment** card, connect an AWS account if your agent calls Amazon Bedrock (as the sample agent does), or add a provider key there for a non-Bedrock model. Local development uses these credentials even after you sign in to your space in step 9.
@@ -206 +206 @@ These are the views an operator opens first. Work through them before you open a
-  1. Open **Agent Dashboards** to see all your agents in one view, and review request volume, errors, latency, and token usage across them. Token usage is the signal that drives cost. For what each signal tells you, see [Monitor agents in the web UI](./omni-monitor-agents-in-the-web-ui.html).
+  1. Open **Agent overview** to see all your agents in one view, and review request volume, errors, latency, and token usage across them. Token usage is the signal that drives cost. For what each signal tells you, see [Monitor agents in the web UI](./omni-monitor-agents-in-the-web-ui.html).
@@ -240 +240 @@ A failure like this raises no alarm. The run succeeded, so nothing routed you to
-  2. Add them to a new dataset named `scope-regression`. For your own agent, name it after the failure you are fixing. Later steps say `scope-regression`, so substitute your name as you go.
+  2. Add them to a new dataset named `scope_regression`. For your own agent, name it after the failure you are fixing. Later steps say `scope_regression`, so substitute your name as you go.
@@ -249 +249 @@ The dataset is the handoff artifact. It carries the real failing inputs, so the
-**Checkpoint:** `scope-regression` exists in your space with the failing test cases.
+**Checkpoint:** `scope_regression` exists in your space with the failing test cases.
@@ -255 +255 @@ Steps 9–13 happen back in your editor, and they move the extension through bot
-Step 9 connects to your space, which switches the extension to cloud mode. Step 10 downloads the dataset from the space. Step 12 switches back to local mode to run the fix. The status bar shows which mode is active throughout.
+Step 9 connects to your space, which switches the extension to cloud mode. Step 10 downloads the dataset from the space. Step 12 switches back to local mode to run the fix. The **Data source** control at the top of the extension sidebar shows which mode is active throughout.
@@ -269 +269 @@ The extension has two independent sign-ins, and this step is about the first one
-  * Choose **Sign In**. Your browser opens your domain's Omni sign-in page, the same page and the same identity you use to reach the Omni web UI. Complete sign-in there.
+  * Choose **Sign in**. Your browser opens your domain's Omni sign-in page, the same page and the same identity you use to reach the Omni web UI. Complete sign-in there.
@@ -275 +275 @@ The extension has two independent sign-ins, and this step is about the first one
-  * Cloud mode turns on by itself once a space is selected. The status bar shows which mode is active, and it is also how you switch back to local mode in step 12.
+  * Cloud mode turns on by itself once a space is selected. The **Data source** control at the top of the sidebar shows which mode is active, and it is also how you switch back to local mode in step 12.
@@ -277 +277 @@ The extension has two independent sign-ins, and this step is about the first one
-  * Leave **Use this account for local dev** selected. Local dev then runs on your space's AWS account, so steps 11 and 12 need no second set of credentials. Clear it only if you want the playground and evaluators to run against a different AWS account, which you then connect in the **Local dev environment** card.
+  * Clear **Use this account for local dev**. The playground and evaluators in steps 11 and 12 run on your machine with the AWS account connected in the **Local dev environment** card, not with your space sign-in. If you did not connect one in step 4, choose **Connect** in that card, choose **Automatic** for **Method** , choose your Region, and then choose **Connect**.
@@ -294 +294 @@ This step runs in cloud mode, which step 9 turned on: the **Datasets** panel lis
-  1. Open **Datasets** in the extension. `scope-regression` appears in the list.
+  1. Open **Datasets** in the extension. `scope_regression` appears in the list.
@@ -303 +303 @@ The download creates an independent local copy of the dataset. There is no file
-**Checkpoint:** the download completes, and `scope-regression` exists as a local dataset. You see it listed when you switch to local mode in step 12.
+**Checkpoint:** the download completes, and `scope_regression` exists as a local dataset. You see it listed when you switch to local mode in step 12.
@@ -309 +309 @@ The failure is a judgment about meaning, so create a custom model-as-judge evalu
-  1. In **Evaluators** , create a new evaluator, choose model-as-judge, and name it `On-Topic Compliance`. For your own agent, name it after the rule it enforces. Later steps say `On-Topic Compliance`.
+  1. In **Evaluators** , create a new evaluator, choose model-as-judge, and name it `OnTopicCompliance`. For your own agent, name it after the rule it enforces. Later steps say `OnTopicCompliance`.
@@ -349 +349 @@ For your own agent, keep this shape and replace the specifics: say what the agen
-  1. Switch the extension back to local mode. Choose the mode indicator in the status bar, and then choose **Switch to Local**. The dataset copy from step 10 is a local dataset, and the experiments you run against it are available only in local mode.
+  1. Switch the extension back to local mode. Under **Data source** at the top of the extension sidebar, choose **Local**. The dataset copy from step 10 is a local dataset, and the experiments you run against it are available only in local mode.
@@ -368 +368 @@ For your own agent, write the same instruction in its terms: state what it may d
-  6. Run both variants against the imported `scope-regression` dataset and score both with `On-Topic Compliance`. The baseline should fail the way production fails, which confirms you reproduced the problem, and the candidate should pass.
+  6. Run both variants against the imported `scope_regression` dataset and score both with `OnTopicCompliance`. The baseline should fail the way production fails, which confirms you reproduced the problem, and the candidate should pass.
@@ -412 +412 @@ An online evaluation scores sampled live traffic continuously, so that this regr
-  2. Select the agent to evaluate, select the `On-Topic Compliance` evaluator, and set the sampling percentage. Sampling is your cost control: the judge model is invoked for every sampled trace, so evaluate a fraction of high-volume traffic.
+  2. Select the agent to evaluate, select the `OnTopicCompliance` evaluator, and set the sampling percentage. Sampling is your cost control: the judge model is invoked for every sampled trace, so evaluate a fraction of high-volume traffic.
@@ -423 +423 @@ An online evaluation scores sampled live traffic continuously, so that this regr
-**Checkpoint:** the online evaluation is enabled, and `On-Topic Compliance` trends at 1 on live traffic.
+**Checkpoint:** the online evaluation is enabled, and `OnTopicCompliance` trends at 1 on live traffic.
@@ -429 +429 @@ Each artifact in the loop does one job, and each keeps working after the fix shi
-  * **The dataset** `scope-regression` pins down the exact failing scenarios. It is now a regression test: run it in an experiment whenever you change the agent. See [Datasets](./omni-agents-datasets.html).
+  * **The dataset** `scope_regression` pins down the exact failing scenarios. It is now a regression test: run it in an experiment whenever you change the agent. See [Datasets](./omni-agents-datasets.html).
@@ -431 +431 @@ Each artifact in the loop does one job, and each keeps working after the fix shi
-  * **The evaluator** `On-Topic Compliance` defines what "broken" means identically wherever it runs: in the playground during the fix, and on live traffic afterward.
+  * **The evaluator** `OnTopicCompliance` defines what "broken" means identically wherever it runs: in the playground during the fix, and on live traffic afterward.