Create Pattern-Based Exclusions from Ignored Findings
Objective
Ignored findings are often suppressed one at a time. Each manual ignore is a judgment call with no shared scope, no justification on record, and no expiration date. Few of them get revisited. This playbook shows you how to audit those accumulated decisions for patterns and convert the repeating ones into properly scoped exclusions with documented business context.
Step 1: Audit Your Ignored Findings for Patterns
Before creating any exclusions, understand what is already being ignored and why. The Ignored tab is a record of individual decisions. Group it by policy and the noise patterns become visible.
To audit ignored findings:
- Navigate to Risks > Cloud Findings.
- Click Ignored.
-
Click Group by > Policy.
What a high count tells you: A policy with 20 or more ignored findings means either an existing exclusion is doing its job, or findings are being manually suppressed one at a time. Both are worth reviewing. - For each policy group with a high ignored count, apply the following filters one at a time:
a. Filter by Account to check whether the noise is concentrated in one environment, such as a Dev or QA account.
b. Filter by Sub-Status to see the breakdown of By design, Exception, and False positive reasons.
c. Filter by Resource to look for naming patterns, such as resources whose names start with lab-, test-, qa-, or sprint-. - Click the Export button at the top-right corner of the table to save the filtered view to CSV.
What makes a strong exclusion candidate: Look for these signals across the filtered results:
- The same policy appears repeatedly across multiple resources.
- The resources follow a consistent naming pattern or share a tag, such as env:dev.
- The findings are concentrated in one account or folder, not spread across production.
If you see a policy with 50 ignored findings in a single Dev account, all on resources named lab-*, that is one exclusion worth building.
Step 2: Build Pattern-Based Exclusions
Once you identify a repeating pattern, convert it into an exclusion from the Exclusions page. The Exclusions page gives you full control over scope, resource pattern matching, expiration, and business justification. Exclusions created directly from the Findings page are faster but offer fewer configuration options and work best for one-off suppressions.
To create a pattern-based exclusion:
- Navigate to Policies > Exclusions.
- In the Organization drop-down box, select the specific account or folder where the pattern occurs. Do not scope to the full organization unless the pattern genuinely applies everywhere.
- Click Add Exclusion. The account scope cannot be changed after you start.
- In the Name box, type a name that describes the business context, not the finding type. For example: Dev1 publicly accessible EC2 instances — QA testing.
- In the Description box, type the justification. For example: Dev account intentionally exposes EC2 instances for load testing. Reviewed and confirmed not applicable to production. Owner: QA team.
-
Under Action, select Ignore finding and choose a Reason:
- By design: The configuration is intentional and required for functionality.
- Exception: The finding is accepted temporarily while remediation is in progress.
- False positive: The finding was verified as incorrect or irrelevant.
The reason becomes the sub-status displayed for every finding this exclusion covers.
- Under Target Findings, set the Type to Cloud.
-
Under Policies, select Specific policies and choose the relevant policy.
Note: Keep each exclusion scoped to one policy. Mixing unrelated policies in a single exclusion creates coverage overlap that is difficult to audit and harder to justify to stakeholders. -
Under Target Resources, select Specific resources and type a Name pattern using wildcards:
- Asterisk * matches zero or more characters.
- Question mark ? matches a single character.
For example, lab-* targets all resources whose names start with lab-. To also catch a manually named instance that does not follow the convention, add a second condition: OR resource name equals JohnD.
Important: Test your pattern in the Findings Preview panel before saving. A pattern like prod-* on a broadly scoped exclusion will silently exclude newly created production resources that share the same naming prefix. The preview panel shows all currently targeted open findings. - In Expiration, select Never for permanent configurations, or select Expiration date for accepted risks with a known end date.
- Review the Findings Preview panel to confirm the scope matches your intent.
- Click Save.
After saving, the exclusion takes effect in the next scan cycle. Targeted findings move from Open to Ignored with the reason you selected as their sub-status.
For more information about reviewing and retiring exclusions over time, see Audit and Maintain Your Exclusions.