PolicyCortex and CMMC: How Policy-as-Code Automates Evidence Collection Across 110 Practices
CMMC COMPLIANCE

PolicyCortex and CMMC: How Policy-as-Code Automates Evidence Collection Across 110 Practices

How PolicyCortex maps to every CMMC Level 2 practice area, turning manual evidence gathering into continuous, automated compliance.

Leonard EsereMay 8, 2026

CMMC Level 2 requires implementation of all 110 security practices from NIST SP 800-171 Rev 2, spread across 14 control families. For most defense contractors, the hardest part is not understanding what the controls require. It is proving that the controls are implemented, operating effectively, and documented in a way that satisfies a C3PAO assessor. That evidence burden is where most preparation efforts stall. PolicyCortex was built to solve exactly this problem by treating compliance requirements as executable code rather than static documentation.

The Evidence Problem in CMMC

A C3PAO assessment is fundamentally an evidence review. Assessors examine three things for each practice: whether the control is documented in your System Security Plan, whether the control is implemented in your environment, and whether you have evidence that the control is operating as described. The gap between "we do this" and "here is the proof we do this" is where contractors fail.

Traditional compliance preparation treats evidence collection as a project. Someone spends weeks taking screenshots, exporting configurations, writing narratives, and organizing artifacts into folders mapped to control numbers. By the time the evidence package is assembled, half of it is already stale because configurations have changed, users have been added or removed, and policies have been updated. The assessor sees evidence that does not match the current state of the environment, and the finding count climbs.

The Stale Evidence Trap

In our experience, contractors who assemble evidence manually see 30-40% of their artifacts become outdated within 60 days of collection. When the C3PAO arrives three months later, the evidence package tells a story that no longer matches reality. This is the single most common source of assessment findings.

Policy-as-Code: The Core Concept

Policy-as-code means expressing compliance requirements as machine-readable, executable rules rather than prose documents. Instead of a policy document that says "all users must authenticate with multi-factor authentication," a policy-as-code rule checks whether MFA is enforced in your identity provider, evaluates the current state, and records the result as evidence. The policy, the check, and the evidence are all part of the same automated pipeline.

PolicyCortex implements this concept across all 14 NIST 800-171 control families. Each of the 110 practices has a corresponding set of policy-as-code rules that define what "implemented" looks like in your specific environment, whether that environment runs on Azure GCC High, AWS GovCloud, or a hybrid on-premises architecture. The rules are not generic templates. They are calibrated to your technology stack and your CUI boundary.

Mapping PolicyCortex to the 14 Control Families

Here is how PolicyCortex addresses each NIST 800-171 control family with policy-as-code automation:

PolicyCortex Coverage by Control Family

Control FamilyPracticesPolicyCortex Automation
Access Control (AC)22RBAC validation, CAP enforcement, session controls
Awareness & Training (AT)3Training completion tracking, role-based assignment
Audit & Accountability (AU)9Log retention validation, SIEM integration, review scheduling
Configuration Mgmt (CM)9Baseline drift detection, change tracking, software inventory
Identification & Auth (IA)11MFA enforcement, password policy checks, credential management
Incident Response (IR)3IR plan validation, incident tracking, reporting automation
Maintenance (MA)6Patch compliance tracking, maintenance logging
Media Protection (MP)9DLP policy enforcement, encryption validation, media tracking
Personnel Security (PS)2Screening status tracking, termination workflow validation
Physical Protection (PE)6Access log integration, visitor tracking, facility monitoring
Risk Assessment (RA)3Vulnerability scan scheduling, risk scoring, remediation tracking
Security Assessment (CA)4Continuous monitoring, POA&M management, assessment scheduling
System & Comm Protection (SC)16Encryption validation, boundary protection, network segmentation
System & Info Integrity (SI)7Flaw remediation tracking, malware protection, alert monitoring

How Evidence Collection Actually Works

PolicyCortex collects evidence through three mechanisms: API-based configuration pulls, agent-based endpoint telemetry, and integration-based log aggregation. Here is what that looks like in practice for a typical CMMC engagement:

Evidence Collection Pipeline

1

Environment Discovery

PolicyCortex connects to your cloud tenant (Azure, AWS, or hybrid) via read-only service principals and inventories all resources within the CUI boundary. This produces the asset inventory required by CM-8.

2

Policy Evaluation

Each policy-as-code rule runs against the discovered environment state. For example, the AC-2 rule checks whether all user accounts have appropriate role assignments and whether inactive accounts exceed the defined threshold.

3

Evidence Artifact Generation

Each evaluation produces a timestamped evidence artifact: the rule that was evaluated, the data it examined, the result (pass, fail, or partial), and a hash for integrity verification. These artifacts are stored in an immutable evidence vault.

4

SSP Narrative Update

When evidence shows a control state change, the corresponding SSP section is flagged for review. PolicyCortex generates draft narrative updates based on the new evidence, keeping the SSP synchronized with reality.

5

Assessment Package Assembly

When assessment time arrives, PolicyCortex assembles the complete evidence package organized by control family, with cross-references between the SSP narrative, the policy-as-code rule, and the evidence artifact for each practice.

Deep Dive: Access Control Family Automation

Access Control is the largest control family with 22 practices, and it is also the family where PolicyCortex delivers the most automation value. Consider practice 3.1.1, which requires limiting system access to authorized users. In a traditional compliance approach, you would document your access control policy, take screenshots of your user directory, and write a narrative explaining how access is granted and revoked.

With PolicyCortex, the policy-as-code rule for 3.1.1 continuously evaluates your identity provider against defined criteria: Are all accounts tied to active employees? Are there orphaned accounts from terminated users? Do service accounts have appropriate scope limitations? Are privileged accounts separated from standard user accounts? Each check runs on a configurable schedule (daily by default), and every evaluation result is stored as evidence. When the assessor asks "how do you limit system access to authorized users," you do not hand them a screenshot from three months ago. You show them a continuous compliance record with daily evaluations going back to the start of your preparation.

Handling the Non-Technical Controls

Not all 110 practices can be fully automated. Personnel Security (PS), Physical Protection (PE), and parts of Awareness and Training (AT) involve human processes that cannot be reduced to API calls. PolicyCortex handles these through workflow automation rather than technical scanning. For PS-related practices, PolicyCortex integrates with HR systems to track screening status and trigger access revocation workflows when personnel changes occur. For PE practices, PolicyCortex provides checklists and evidence upload workflows that prompt facility managers to submit physical access logs and visitor records on a defined schedule.

The goal is not to eliminate human involvement in these controls. It is to ensure that the evidence of human involvement is captured consistently and mapped to the correct practice number. An assessor does not care whether the evidence was generated by an API call or uploaded by a facility manager. They care that it exists, that it is current, and that it demonstrates the control is operating.

Automation Coverage by Practice Type

72
Fully Automated
API-based evidence collection
27
Workflow-Assisted
Guided evidence capture
11
Manual with Tracking
Scheduled reminders and upload

POA&M Management and Remediation Tracking

When PolicyCortex identifies a practice that is not fully implemented, it automatically generates a Plan of Action and Milestones (POA&M) entry with the practice number, the specific finding, a recommended remediation path, and a target completion date based on the complexity of the fix. This is critical because CMMC allows conditional certification with open POA&M items, but only if those items are documented with credible remediation plans and timelines.

The POA&M is not a static spreadsheet. PolicyCortex tracks remediation progress against each item and automatically closes POA&M entries when the corresponding policy-as-code rule begins passing. This gives you a verifiable remediation record that shows assessors not just what was broken, but when it was fixed and how you verified the fix.

Continuous Monitoring After Certification

CMMC certification is not a one-time event. Certified organizations must maintain their security posture and submit annual affirmations. PolicyCortex continues running after certification, providing the continuous monitoring capability that NIST 800-171 practice 3.12.3 requires. If a configuration drifts out of compliance, if a new vulnerability is discovered, or if an access control change introduces a gap, PolicyCortex flags it immediately and generates the evidence trail that demonstrates your ongoing compliance posture.

Annual Affirmation Support

When your annual affirmation window arrives, PolicyCortex generates a compliance summary report that documents your posture across all 110 practices for the preceding 12 months. This report, combined with the continuous evidence record, provides the substantiation your senior official needs to sign the affirmation with confidence.

The Bottom Line

CMMC Level 2 compliance is an evidence problem as much as it is a security problem. You can implement every control perfectly, but if you cannot prove it to an assessor with current, organized, and verifiable evidence, you will not pass. PolicyCortex turns evidence collection from a periodic project into a continuous, automated pipeline that keeps your SSP, your evidence package, and your POA&M synchronized with the actual state of your environment.

For contractors preparing for a C3PAO assessment, the difference between manual evidence collection and PolicyCortex automation is typically four months of preparation time and a significantly lower finding count. The policy-as-code approach does not just make compliance faster. It makes it more accurate, because the evidence always reflects what your environment actually does, not what someone remembered to screenshot three months ago.

See PolicyCortex Map to Your CMMC Practices

Schedule a demo to see how PolicyCortex automates evidence collection across all 110 NIST 800-171 practices for your specific environment.

Request a PolicyCortex Demo
LE

Leonard Esere

Founder & CEO, AeoliTech

Leonard has guided dozens of defense contractors through CMMC preparation, drawing on deep experience with NIST frameworks at MITRE and LANL. He leads AeoliTech's mission to make compliance achievable through policy-as-code automation.