Evaluate-STIG and STIG Viewer are not competitors, and the question of which one to use has a short answer: both, in that order. Evaluate-STIG is a PowerShell-based scanner maintained by the Navy (NSWC Crane) that runs the automatable STIG checks against a host and writes the results into a checklist file. STIG Viewer is the DISA-maintained desktop application that opens that checklist file, lets you review and answer the checks a scanner cannot decide, and produces the finished .ckl or .cklb that goes into eMASS. One produces the draft; the other is where you finish it. The ISSO who tries to build checklists in STIG Viewer alone is hand-answering four hundred Windows Server checks per host, and the one who trusts Evaluate-STIG output without opening it is submitting “Not Reviewed” as a compliance status.
Below: what each tool does, the ckl versus cklb question (STIG Viewer 3 changed things), the import path into eMASS, and the manual checks that stay yours no matter how good the automation gets.
What Evaluate-STIG does (and does not do)
Evaluate-STIG is a script package you run on or against a target host, with local administrator rights, that evaluates each Vulnerability ID (V-number) in the applicable STIGs and records Open, NotAFinding, Not_Applicable, or Not_Reviewed along with the finding details and the evidence it collected. It ships with modules for the operating systems and applications DISA publishes STIGs for, and it automatically detects which STIGs apply to the host (Windows Server 2022, Microsoft Defender, .NET Framework, Edge, IIS, SQL Server, and so on). On a typical Windows member server it will confidently answer somewhere in the range of 70 to 85 percent of checks, with the rest handed back to you as Not_Reviewed because the check requires reading a policy document, interviewing a human, or looking at something the script cannot see.
Three things it does that matter to an ISSO. It applies answer files, which are XML files where you pre-record justifications for checks that are consistently Not Applicable or consistently Open with a documented mitigation on your systems, so the same fifteen findings do not need retyping across forty hosts every quarter. It produces a consistent output structure per host, per STIG, per run, dated, which is exactly what continuous monitoring evidence needs to look like. And it is fast: a full run on a Windows server is minutes, not the afternoon a manual pass takes.
Things it does not do. It does not fix anything. It does not decide whether an Open finding is actually mitigated by something outside the host. It does not know your organization’s approved deviations unless you put them in an answer file. And it will happily report Not_Reviewed on checks it cannot evaluate, which is not a compliance status the SCA accepts. Evaluate-STIG output is a starting point that removes the tedious part; it does not remove the ISSO.
What STIG Viewer does
STIG Viewer is the reference reader and checklist editor. You import the STIG (the XCCDF benchmark zip from DISA), it displays every rule with its check text and fix text, and you create or open a checklist that records a status and comment for every rule against a specific asset. It also imports SCAP results from the SCAP Compliance Checker (SCC), which is the DISA-authored automated path, and it is the tool the assessor opens when they want to see your answers.
The current major version is STIG Viewer 3, which replaced the Java-based 2.x line. The visible changes: a cleaner interface, an asset-centric view where one asset can hold multiple STIG checklists, and a new checklist file format.
ckl vs cklb: the format question
STIG Viewer 2.x wrote checklists as .ckl files, which are XML. STIG Viewer 3 writes .cklb files, which are JSON. Viewer 3 can open and import .ckl files and can export back to .ckl for compatibility, and Evaluate-STIG can output both formats (it has been able to produce .cklb since the Viewer 3 transition, and the shops I have worked in still produce .ckl by default because their eMASS import process was built around it).
Which one you should use depends entirely on what the downstream consumer accepts. As of 2026, the rule I follow:
- Check what your eMASS instance accepts for STIG checklist import. Import support for .cklb has been rolling out across instances, but do not assume; test one file before you convert two hundred.
- Check what your SCA wants delivered. Some assessment teams have automation built on .ckl parsing and will ask for it explicitly.
- Check what your organization’s STIG Manager instance (if you have one) ingests. STIG Manager is the DISA-supported web application for aggregating checklists across many assets, and it is increasingly the middle layer between the scanners and eMASS.
- Keep the Evaluate-STIG raw output regardless of format. The results folder with the evidence it collected is what you will want when someone asks in nine months why a check was marked NotAFinding.
If nobody downstream cares, produce .cklb, because that is the direction the tooling is going. If anyone downstream cares, produce what they care about. File format is not the hill.
The workflow, start to finish
Here is the sequence I run on a system with a few dozen hosts, quarterly for ConMon and in full before any assessment. It assumes you have already applied the STIGs as part of implementation and are now producing the evidence that you did.
- Confirm the STIG versions. Download the current quarterly STIG library from DISA and confirm Evaluate-STIG has been updated to match. Running a scanner built for V2R1 against a benchmark that is now V2R4 gives you a checklist the SCA will reject on the first page.
- Update answer files. Review the previous quarter’s deviations: are they still valid, are the mitigations still in place, has anything been closed. Retire stale ones. An answer file that marks a check Not Applicable because “this is a virtual machine” on a host that has since moved to bare metal is how you get a CAT II turned into a CAT I with an integrity question attached.
- Run Evaluate-STIG. Locally with an elevated PowerShell session, or remotely against a host list. Output to a dated folder per host. Do not overwrite last quarter’s results; you need the trend.
- Open every checklist in STIG Viewer. Filter to Not_Reviewed. Answer each one with a status and a comment that says what you looked at. “Reviewed GPO, setting confirmed” is a comment. “N/A” is not.
- Filter to Open. For each Open finding, decide: fix it, document a mitigation and leave it Open with a POA&M, or determine it is actually Not Applicable with a written reason. CAT I findings get handled first; the process for closing CAT I findings without breaking the system is its own discipline.
- Fill in the asset data. Hostname, IP, MAC, FQDN, role, and technology area. eMASS matches checklist imports to assets in the Hardware baseline on these fields, and a blank hostname is an orphaned checklist.
- Save, name consistently, archive. Something like HOSTNAME_STIGNAME_V2R4_2026Q3.cklb. You will thank yourself at reauthorization.
- Import to eMASS. Covered below.
Getting the results into eMASS
eMASS consumes STIG checklists through the Asset Manager module (labels vary by instance). You upload the .ckl or .cklb, eMASS associates it with an asset in your Hardware baseline, and the individual check results roll up to the CCIs and assessment procedures mapped to each STIG rule. That roll-up is what updates the control status you see in the Controls tab, and it is why a control like CM-6 can show non-compliant even though you swear you configured everything: one Open check on one host, mapped to one CCI, flips it.
Three practical notes on the import. First, the asset must exist in the Hardware baseline before the checklist will attach cleanly; import the baseline, then the checklists. Second, eMASS treats Not_Reviewed as not compliant, so every one of those you left behind in STIG Viewer is now a finding in the package. Third, checklist results and manual test results entered against CCIs coexist, and the manual entry does not override the checklist result. If you have a documented mitigation for an Open STIG finding, it goes in the POA&M and in the checklist comment, not in a test result that pretends the finding is closed.
If your organization runs STIG Manager, the path becomes Evaluate-STIG to STIG Manager to eMASS, with STIG Manager doing the aggregation, review workflow, and export. That is a better model for anything over a few dozen hosts, because it gives you one place to see every asset’s status instead of a folder of four hundred files.
The manual checks you cannot automate away
Every STIG has checks that require a human, and Evaluate-STIG will flag them as Not_Reviewed with a note. These cluster into predictable categories, and I keep a standing list per technology so I am not rediscovering them every quarter.
- Documentation checks. “Verify the organization has a documented process for…” The evidence is a policy, an SOP, or an SSP section, and the checklist comment should cite it by name and date.
- Account and role reviews. Checks that ask whether every member of the local Administrators group is authorized. The scanner can list the group; only you know who is supposed to be in it.
- Physical and environmental items. Server room access, BIOS passwords set at the console, removable media controls. Nobody scripts a walk to the data center.
- Interview checks. “Interview the system administrator to determine whether…” Write down who you asked, when, and what they said.
- Checks dependent on architecture. Whether a required protection is provided by another component (a network device, an enterprise service). Often Not Applicable with an inheritance justification, but it has to be written.
A useful habit: for each technology, keep the manual-check answers in the answer file with a review date, and refresh them quarterly. That converts a recurring two-hour chore into a fifteen-minute confirmation, and it produces a paper trail that shows the SCA you reviewed rather than copy-pasted.
What the SCA is actually looking for
When the assessor opens your checklists, they are checking four things: the STIG version matches the current release, the date is inside the assessment window, there are zero Not_Reviewed, and every Open and Not Applicable has a comment that a stranger could understand. Get those four right and the checklist review is quick. Miss any of them and the assessor starts spot-checking your NotAFindings against the live host, which is the part of the day nobody enjoys. The full STIG compliance walkthrough covers what that spot-check looks like from the receiving end. And pair whichever scanner you run (Evaluate-STIG, or the DISA SCAP Compliance Checker if your shop standardized on it) with ACAS vulnerability scanning, because STIG compliance and patch compliance are different questions, and the SCA will ask both.
Every Open finding that survives this workflow becomes a POA&M item with a milestone, a resource, and a date, and the checklists you archived each quarter are the evidence that the milestone was real. If you are tracking those items in a spreadsheet you rebuilt from memory after the last assessment, the POA&M Tracker is the one I built to keep STIG findings, mitigations, and dates in one place between scans.


Leave a Reply