Join our Newsletter — 33% off our NHI Course
Home› Guides› SCAP and Security Content Automation Guide: XCCDF, OVAL…
Guide Endpoint Security

SCAP and Security Content Automation Guide: XCCDF, OVAL and Data Streams Explained

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 27 October 2026 13 min read
On this page

Configuration and vulnerability checks written as prose in a PDF benchmark cannot be run, compared or audited at scale, and every team that hand-translates them into scripts produces a slightly different answer. The Security Content Automation Protocol (SCAP) is NIST's suite of specifications that expresses checklists, checks, platforms and results in machine-readable XML so that any conforming tool can evaluate the same content and report in the same format. This guide explains what each SCAP component does, how a data stream is assembled and evaluated, the current state of the standard in 2026 (SCAP 1.3, work on 1.4, the end of the NIST validation programme and the archived OVAL repository), and how to build a sustainable SCAP practice. It is for endpoint, platform and compliance engineers. Hands-on scanning with the oscap tool is covered in the OpenSCAP and Linux Compliance Scanning Guide; this page sits one level up, at the standards and content layer.

Key takeaways

  • SCAP is a family, not a single format. XCCDF describes the checklist, OVAL and OCIL describe the checks, CPE and CCE name platforms and settings, and ARF carries the results.
  • SCAP 1.3 is the published specification. NIST SP 800-126 Rev. 3 (February 2018) defines it, and NIST states that SCAP is maintained through the development of SCAP 1.4.
  • The ecosystem is changing. NIST has concluded the SCAP Validation Program (test reports were accepted only until 2 September 2025) and the CIS OVAL Repository was archived read-only on 10 January 2025.
  • Data streams are the unit to deploy. A source data stream bundles the benchmark, checks and dictionaries into one file, which is what scanners such as OpenSCAP expect.
  • Content quality decides outcomes. Most SCAP failures come from stale, untailored or platform-mismatched content, not from the scanner.

What SCAP is and why it exists

NIST SP 800-126 Rev. 3 describes SCAP as a suite of specifications that standardise the format and nomenclature for communicating software flaw and security configuration information, between machines and between machines and people. Before SCAP, a hardening checklist such as a government baseline existed as a document, and every scanner vendor wrote its own checks. Results from two tools were not comparable, and an auditor could not easily see which check corresponded to which requirement.

SCAP separates four concerns so each can be standardised and reused:

  • What to check: a checklist with rules, grouped into profiles (XCCDF).
  • How to check: machine-evaluable logic against system state (OVAL) or structured questions for things a machine cannot see (OCIL).
  • What it applies to: standard names for platforms and configuration items (CPE, CCE) and for vulnerabilities (CVE, scored with CVSS).
  • What was found: standard result and asset reporting formats (XCCDF results, ARF, Asset Identification).

NIST's National Checklist Program (NCP), now described in SP 800-70 Rev. 5 (May 2026, superseding Rev. 4), collects security configuration checklists, many of which are published as SCAP content, through the National Checklist Repository at checklists.nist.gov.

The SCAP components

ComponentTypeWhat it doesPractical note
XCCDF (Extensible Configuration Checklist Description Format)Checklist languageDefines a benchmark of rules, groups, values and profiles, and records results per ruleThe profile is what you select at scan time; tailoring changes values and rule selection
OVAL (Open Vulnerability and Assessment Language)Check languageExpresses tests of system state: files, registry keys, packages, kernel parameters, servicesDefinitions are grouped into classes such as compliance, vulnerability, inventory and patch
OCIL (Open Checklist Interactive Language)Check languageStructured questionnaires for checks needing human inputUse for procedural controls a scanner cannot observe
CPE (Common Platform Enumeration)EnumerationStandard names for platforms, operating systems and applications, plus an applicability languageDrives "notapplicable" results when a rule does not fit the target
CCE (Common Configuration Enumeration)EnumerationIdentifiers for individual configuration settingsLets one setting be traced across different benchmarks
CVE and CVSSEnumeration and metricVulnerability identifiers and severity scoringUsed by vulnerability-class OVAL content; see the CVE, CWE and CPE Guide
ARF (Asset Reporting Format) and Asset IdentificationReportingPackages results with the asset they describeThe format to keep as audit evidence and feed to aggregation tools
TMSAD (Trust Model for Security Automation Data)IntegritySignatures and trust for SCAP contentSupports verifying that content came from its publisher

NIST lists AI, ARF, CCE, CPE (including the CPE 2.3 specifications and work on CPE 3.0), OVAL, OCIL, TMSAD and XCCDF as SCAP component specifications on its SCAP project page.

How content is assembled and evaluated

From rule to result

  1. A rule in the XCCDF benchmark states a requirement (for example, that password authentication is disabled for a remote login service), with references to CCE identifiers and to the controls it supports.
  2. The rule's check points to an OVAL definition that tests the relevant configuration file or setting, or to an OCIL question.
  3. Applicability is expressed with CPE: the rule only runs on platforms it names, so a Linux rule on Windows returns notapplicable rather than fail.
  4. A profile selects which rules run and which values apply (a 14-character minimum password length in one profile, 15 in another).
  5. The scanner evaluates the selected checks and writes XCCDF results per rule (pass, fail, notapplicable, notchecked, error and similar states), usually wrapped in ARF with asset details.
  6. Optional remediation content attached to rules (shell, Ansible or other fix types) can be generated or applied by the tool.

Data streams

A SCAP source data stream is a single XML file that bundles the XCCDF benchmark, the OVAL and OCIL check content and the CPE dictionary. The ComplianceAsCode project, which builds widely used Linux content and began in 2011 as SCAP Security Guide before being renamed in September 2018, recommends data streams for general use because they contain everything needed to evaluate and remediate a machine. The same project generates Ansible playbooks and Bash scripts from the same rules, and CEL content evaluated by the Kubernetes compliance-operator for Kubernetes and OpenShift.

Tailoring

Tailoring files let you change rule selection and values without editing the vendor's benchmark. Keep the upstream data stream unchanged and version your tailoring file separately, so that content updates can be taken without merging local edits. The OpenSCAP and Linux Compliance Scanning Guide shows tailoring in practice.

The state of SCAP in 2026

SCAP is mature but its supporting ecosystem has shifted, and anyone building a programme now should plan for that:

ItemStatus (checked 27 October 2026)What it means for you
SCAP 1.3 (SP 800-126 Rev. 3)Final, February 2018, with annex SP 800-126AThe version most tools and content target today
SCAP 1.4In development; NIST invites comments at [email protected]Watch for a new SP 800-126 revision before changing content pipelines
SCAP v2NIST project for a major update with broader scope; no milestones published on the project pageDo not design around it yet
SCAP Validation ProgramArchived; phased conclusion; NVLAP accepted test reports only until 2 September 2025 and takes no new accreditations or renewals"SCAP validated" claims will age; test tools against your own content instead
CIS OVAL RepositoryArchived read-only on 10 January 2025 (it continued MITRE's repository, which MITRE ran until 31 July 2015)Use vendor and project-maintained OVAL feeds rather than the central repository
National Checklist ProgramActive; SP 800-70 Rev. 5 published May 2026Still the federal index of checklists, including SCAP content

Two practical consequences follow. First, content provenance now matters more than tool certification: take benchmarks from the publisher (operating system vendors, ComplianceAsCode, DISA, CIS) and track their release versions. Second, many organisations now pair SCAP scanning with configuration expressed in other forms, such as policy-as-code in configuration management or OSCAL for control documentation. See the OSCAL Guide.

Choosing content and tools

  • Linux fleets: ComplianceAsCode data streams cover Red Hat Enterprise Linux, Fedora, Ubuntu, Debian and SUSE Linux Enterprise Server, with CIS, PCI DSS and government profiles, scanned with oscap, oscap-ssh or SCAP Workbench.
  • US federal and defence: DISA STIG benchmarks published as SCAP content are the usual starting point. See the DISA STIG Guide.
  • CIS-aligned programmes: check which machine-readable formats your CIS membership tier provides before planning to scan with SCAP tools. See the CIS Benchmarks Guide.
  • Windows-heavy estates: SCAP scanners exist for Windows, but many teams enforce baselines with Group Policy or Intune and use SCAP scans as independent verification. See the Microsoft Intune Guide.
  • Containers and Kubernetes: scan images in the pipeline with a data stream, and use the compliance-operator CEL content for cluster nodes rather than shelling into nodes.

Small organisations gain most from running one vendor-maintained profile on a schedule and fixing failures; building custom OVAL is rarely worth it. Large organisations benefit from a content pipeline: pinned upstream data streams, version-controlled tailoring, automated test scans on reference images and central ARF collection.

Writing and maintaining custom content

Sooner or later a team needs a check that no upstream benchmark provides: an internal agent that must be installed, a banned package, a required log forwarding setting. Before writing SCAP XML by hand, consider these options in order:

  • Contribute upstream: if the requirement is generic (a common hardening setting), open-source projects such as ComplianceAsCode accept rules written in their own templated source format, which then generates the OVAL, remediation and documentation for every supported platform.
  • Use a project's build system privately: the same templating can be used in a private fork for organisation-specific rules, keeping the generated data stream consistent with upstream structure.
  • Hand-written OVAL: appropriate for a small number of stable checks. Keep each definition small, reference a CCE or internal identifier, and test it against a known-compliant and a known-non-compliant reference host so that both pass and fail paths are proven.
  • OCIL for procedural checks: where the answer comes from a person (for example, whether a manual key ceremony was witnessed), record it as a questionnaire rather than pretending a scanner can verify it.

Whatever the route, treat content as code: peer review, automated test scans in CI against reference images, semantic version numbers on the generated data stream and a changelog that auditors can read. Content that is not tested regularly drifts out of step with the platforms it describes faster than the platforms themselves change.

Where identity fits

Credentialed SCAP scanning needs an account on every target, often with root or local administrator rights, and remote scanners hold those credentials centrally. That account is a high-value non-human identity: restrict it to the scanning hosts, prefer SSH certificates or short-lived credentials over static keys and passwords, limit privileged commands to what checks need, and log its use. SCAP content can also check identity configuration itself, such as SSH daemon settings, password policy and sudo rules. See the SSH Key Management Guide and the Local Administrator Password Solution (LAPS) Guide.

Implementing an SCAP programme

  1. Inventory platforms and map each to an available benchmark and profile; record gaps where no SCAP content exists. See the CIS Control 1 Inventory and Control of Enterprise Assets Guide.
  2. Pin content versions in a repository and verify signatures or checksums on download.
  3. Tailor once, centrally, with documented reasons for every deselected rule or changed value, and owner approval.
  4. Scan in report-only mode on reference images and a pilot ring, compare against known-good builds, and fix false positives through tailoring rather than by ignoring results.
  5. Remediate in rings: generated Ansible or scripts applied to test, then pilot, then production, with a rollback plan (configuration management revert or image rebuild) for each ring.
  6. Collect ARF centrally and keep it as audit evidence, linked to asset IDs and content versions.
  7. Review content quarterly and on each upstream release, rerunning the reference scans before rolling out.

Detection and measurement

  • Pass rate per profile and platform, trended over time, excluding notapplicable results.
  • Count of error and notchecked results: rising numbers usually mean content and platform mismatch or scanner permission problems.
  • Coverage: percentage of in-scope assets scanned in the last 7 or 30 days.
  • Content age: days since the deployed data stream was released upstream.
  • Drift: rules that move from pass to fail between scans, which feed the Configuration Drift and Baseline Enforcement Guide process.
  • Exception count and age for deselected rules, reviewed at each tailoring change.

Framework mapping

FrameworkReferenceHow SCAP supports it
NIST SP 800-53 Rev. 5CM-2 Baseline Configuration; CM-6 Configuration Settings; CM-6(1) Automated Management, Application, and VerificationMachine-readable baselines and automated verification
NIST SP 800-53 Rev. 5RA-5 Vulnerability Monitoring and Scanning; CA-7 Continuous MonitoringVulnerability-class OVAL and recurring scans
NIST SP 800-53 Rev. 5CM-8 System Component InventoryInventory-class OVAL and Asset Identification
NIST CSF 2.0PR.PS-01, ID.RA-01, ID.AM-02, DE.CM-09Configuration management, vulnerability identification, inventory and monitoring
CIS Controls v8.1Control 4 Secure Configuration; Control 7 Continuous Vulnerability ManagementSee the CIS Control 4 Secure Configuration of Enterprise Assets and Software Guide
ISO/IEC 27001:2022Annex A 8.9 Configuration management; 8.8 Management of technical vulnerabilitiesEvidence that configurations and vulnerabilities are checked

Common failure patterns

  • Running the wrong content for the platform. A data stream for one major OS release scanned against another produces floods of notapplicable or error results that look like coverage.
  • Editing the upstream benchmark. Local edits to vendor XML make every upstream update a merge project; use tailoring files.
  • Treating pass rate as security. A profile with many deselected rules can pass at 100%. Report exceptions alongside scores.
  • Relying on the archived OVAL repository for vulnerability definitions after January 2025, which leaves new CVEs undetected.
  • Auto-remediating production first. Generated fixes can disable services or lock out remote access; ring-based rollout and a rollback plan are essential.
  • Over-privileged, unmanaged scan credentials shared across all hosts and never rotated.

Practitioner checklist

  • Map every platform to a maintained SCAP data stream and profile.
  • Pin and verify content versions; track upstream releases.
  • Keep tailoring in version control with owner-approved reasons.
  • Test scans on reference images before rollout and after each content update.
  • Remediate in rings with a documented rollback.
  • Collect ARF results centrally with asset and content version.
  • Replace any dependence on the archived OVAL repository with maintained feeds.
  • Manage scanning credentials as privileged non-human identities.
  • Report pass rate together with exceptions, errors and coverage.
  • Watch NIST for SCAP 1.4 and plan content pipeline changes early.

Standards and references

Related NHI Mgmt Group resources: OpenSCAP and Linux Compliance Scanning Guide · Endpoint Hardening Checklist · DISA STIG Guide · CIS Benchmarks Guide

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 27 October 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.