Regulated Data Boundary · Add-on

End-to-end encrypted CUI handling, with an assessment layer that never sees your content

Sensitive evidence lives in the approved Wickr and retention boundary. Athena’s SaaS manages sanitized workflow, status, control mapping, and approved conclusions.

Athena's SaaS is not FedRAMP authorized. When Regulated Data Boundary controls are enabled, AWS Wickr Premium serves as the FedRAMP Moderate-authorized collaboration component for sensitive communications and evidence exchange. Athena’s SaaS remains outside the regulated-data boundary and receives only sanitized, non-CUI, non-ITAR, or customer-approved information.

Regulated communications and evidence remain within the approved AWS Wickr and customer-controlled retention environment. Athena Fortify-CMMC stores only approved metadata, opaque references, workflow status, responsibilities, and sanitized conclusions.

Is Athena Fortify-CMMC FedRAMP authorized?

No. Athena does not represent Fortify-CMMC as FedRAMP authorized. Athena has intentionally architected CUI and regulated evidence out of its SaaS. For regulated engagements, AWS Wickr is used as the authorized secure-collaboration component only after the applicable offering, region, authorization scope, configuration, and supporting AWS services have been validated. CUI and regulated evidence remain within the approved Wickr and customer-controlled retention environment. Athena manages the CMMC workflow through approved metadata, external references, responsibilities, status, and sanitized conclusions.

This design gives the customer a clear, defensible system boundary: regulated content stays in the approved environment, while Athena manages the work required to prepare for and sustain CMMC assessment readiness.

Athena’s advantage is not a claim that its SaaS is FedRAMP authorized. Athena’s advantage is a deliberately controlled architecture that keeps CUI out of its SaaS while coordinating regulated collaboration through a separately validated AWS Wickr environment. This reduces unnecessary compliance exposure while giving customers a governed, traceable CMMC readiness and evidence-management process.

Where the boundary sits

Regulated content never leaves the customer-controlled AWS boundary on the left. Only sanitized references, mapping, status, and approved conclusions pass the gate into Athena on the right.

How can Athena provide a FedRAMP solution?

It does not, and never claims to. No. Athena does not represent Fortify-CMMC as FedRAMP authorized. Athena has intentionally architected CUI and regulated evidence out of its SaaS. For regulated engagements, AWS Wickr is used as the authorized secure-collaboration component only after the applicable offering, region, authorization scope, configuration, and supporting AWS services have been validated. CUI and regulated evidence remain within the approved Wickr and customer-controlled retention environment. Athena manages the CMMC workflow through approved metadata, external references, responsibilities, status, and sanitized conclusions.

Is this defensible in an assessment?

The C3PAO makes that determination, not Athena. What the arrangement gives an assessor is a narrow, evidenced boundary: regulated content stays in the customer-controlled environment, every release into the SaaS is classified server-side and human-approved, and each refusal and approval is written to an append-only log. The six tests below are what an assessor typically probes, and where the supporting record lives.

1 · Customer-controlled regulated boundary
AWS

Secure collaboration

AWS Wickr (offering recorded per engagement)

Region recorded per engagement

Authorization scope unverified

Controlled retention repository

Separately configured customer component

Stays here, always

  • CUI, FCI, and export-controlled material
  • Message bodies and file contents
  • Original evidence artifacts and retained records
  • Screenshots, diagrams, vulnerability detail

2 · Sanitization gate

Server-side, fail-closed. Blocked entries are refused and logged.

Human certifier plus a second approver on every release.

No return path for content
3 · Athena SaaS — outside the boundary
Not FedRAMP authorized

Receives only

  • Opaque artifact and room references
  • Practice, objective, and control mapping
  • Workflow status, owners, and due dates
  • Sanitized, human-approved conclusions

And does this with it

  • Runs the assessment workflow and scoring
  • Predicts findings, readiness, and certification outcome
  • Builds SSP, POA&M, and assessor deliverables

What an assessor tests — and where the record lives

  1. 1

    Is the system boundary drawn and documented?

    Boundary profile, this architecture view, and the SSP boundary section

  2. 2

    Is the authorization scope of the collaboration component evidenced?

    Offering + region + services-in-scope determination, with source reference and validator

  3. 3

    Can you show regulated content never entered the SaaS?

    Server-side classification gate, fail-closed refusals, and the append-only audit log

  4. 4

    Is each release into the SaaS sanitized and human-approved?

    Sanitization review records with certifier and second approver

  5. 5

    Are the retention repository and its supporting AWS services determined?

    Retention stack determinations with evidence reference and accountable owner

  6. 6

    Is the arrangement kept current, not just set up once?

    Periodic revalidation cadences, annual re-attestation, and currency score

Test 2 is not yet satisfied for this engagement: the authorization scope of the recorded offering and region has not been verified against the applicable AWS services-in-scope listing. Until that determination is on record, no authorization is represented here.

This document describes Athena's use of AWS Wickr as a FedRAMP-authorized component. It does not represent Athena's SaaS or Athena Consulting Group as independently FedRAMP authorized.

One artifact, end to end

The same five steps every piece of regulated evidence follows. Steps 2 and 3 happen entirely inside the customer boundary; only step 4’s approved output is visible to Athena.

  1. 1
    Athena SaaS

    Request raised

    Athena creates an evidence request with the practice, objective, owner, and due date. No content yet.

  2. 2
    Inside boundary

    Delivered in the secure room

    The customer submits the artifact inside the approved collaboration room. The file never touches Athena.

  3. 3
    Inside boundary

    Retained in the repository

    The record lands in the controlled retention repository under customer keys, logging, and schedules.

  4. 4
    At the gate

    Sanitized and certified

    A reviewer records the opaque reference, hash, and a sanitized summary. A second person approves it.

  5. 5
    Athena SaaS

    Athena reasons on the reference

    Only that approved record enters Athena, where it drives scoring, findings, POA&Ms, and deliverables.

Interactive data flow — click any stage

One artifact, six stages. Select a stage to see exactly what changes about the data, which NIST SP 800-171 families apply there, and which records an assessor can pull as proof. Export the whole map — every stage plus the control mapping — as a print-friendly document or PDF for your audit package.

Stages 1–3 · inside the regulated boundary
4 · gate
Stages 5–6 · all Athena ever sees
Return path from Athena into the boundary— there is none. Content only ever moves left to right, once.
Inside the regulated-data boundary
Stage 1 of 6
Owner: Contractor (data owner)

Artifact created by the contractor

The contractor produces or exports the source artifact (config export, log extract, screenshot, policy) and applies its own marking. No Athena component participates.

Data in

Nothing yet — the artifact originates on contractor-controlled systems.

Data out

Original artifact with full CUI/FCI content, marked at the owner’s classification.

Controls that apply here

  • 3.1 Access ControlContractor authorizes who may produce and handle the artifact.
  • 3.8 Media ProtectionMarking and handling of the original media stays with the owner.

Artifacts produced

  • Evidence request record (non-CUI): what was asked for, by whom, when

Never present at this stage

  • Athena has no account, agent, or read path on the originating system

This document describes Athena's use of AWS Wickr as a FedRAMP-authorized component. It does not represent Athena's SaaS or Athena Consulting Group as independently FedRAMP authorized.

Which stage satisfies which control?

Filter by control family, then jump to the exact stage of the flow above that carries it. Families with no stage touchpoint are satisfied at the layer level only.

Family

3.1 Access Control

AC
3 stages

3.3 Audit and Accountability

AU
5 stages

3.4 Configuration Management

CM
2 stages

3.5 Identification and Authentication

IA
1 stage

3.6 Incident Response

IR
1 stage

3.8 Media Protection

MP
2 stages

3.9 Personnel Security

PS
1 stage

3.12 Security Assessment

CA
2 stages

3.13 System and Communications Protection

SC
3 stages

Hosting and Data Residency

Layer allocation only

Allocated in the layer matrix below; no single stage of the artifact flow exercises it.

10 of 10 families shown · 6 stages in the flow

How the systems interact

Three zones, ten systems, one crossing. Select a numbered handoff to see exactly what moves, what is refused, and which side owns the data at that moment.

People

Humans who act on the system. No system-to-system trust.

Control owner

Uploads regulated evidence

Security officer

Approves policy, region, readiness

C3PAO assessor

Receives the audit bundle

Customer regulated boundary

Customer-owned AWS. CUI is created, stored, and retained here — and stays here.

AWS Wickr Premium

E2EE rooms, messages, files

Retention stack

S3 + customer-managed KMS + CloudTrail

Single sanitization gate

Gate

Deny-by-default, two-person certification

No return path CUI messages and files never flow back into the SaaS.

No return path Retained originals stay in the customer boundary, always.

Athena SaaS (assessment engine)

Receives sanitized, non-CUI assessment records only. Never a CUI system of record.

Intake + AI guard

Classifies, refuses, logs

War Room

Objective scoring, forecasts, handoff lane

Package generator

Issuance ledger + SHA-256 manifest

Provenance snapshots

Immutable, defensibility-scored

The single crossing into Athena
step 4 of 9

Who talks to whom

Single sanitization gate → Intake + AI guard

Customer regulated boundary → Athena SaaS (assessment engine)

What moves

Non-CUI assessment record

The only crossing into the SaaS: reference ID, hash, control mapping, sanitized rationale, two approver signatures.

What never moves

Nothing crosses without a recorded classification decision and second-person certification.

Athena is the assessment engine; AWS Wickr and the customer retention stack are the authorized custodians of regulated data. Only step 4 crosses the boundary, and it carries sanitized, non-CUI assessment records.

Where does my data go?

Pick something you would actually send during an assessment. The answer below is computed by the same classification gate that runs server-side — not a diagram someone drew.

Stays inside the customer boundary
Classified CUI
blocked_cui

1 · Where it starts

A message thread discussing how a control is implemented

2 · What the gate decides

This information cannot be entered into Athena's SaaS. Submit and retain it within the approved AWS Wickr and retention boundary. Record only the approved artifact identifier and sanitized status here.

3 · What happens

Stays in the AWS Wickr enclave. Athena records only that the exchange occurred.

4 · Where the original lives

AWS Wickr enclave

Decision recorded under policy boundary-policy-v1. Refusals are logged; the content itself is never copied into the log.

Watch the gate work

Paste sample text — never real regulated content — and see exactly which rule fires. This runs the same pattern set the server uses to refuse an outbound payload.

110/1200 characters. Nothing here is transmitted or stored — screening happens in your browser.

Refused — 2 rules fired

IP address
Vulnerability identifier (CVE)

What a redacted version would look like

Evidence request EV-1042: firewall at [REDACTED: IP address] still exposes the legacy port; see [REDACTED: Vulnerability identifier (CVE)] in the scan.

23 characters removed. In production the payload is refused outright rather than auto-redacted — a human records an approved sanitized summary instead.

Why this is different

Every CMMC program picks one of three models, usually without saying so out loud. Here is the trade each one makes.

Put CUI in the GRC tool

The common shortcut: upload everything and let the platform index it.

  • Where regulated data lives

    Inside a vendor SaaS you do not control

  • What the SaaS can see

    Everything — file bodies, previews, OCR, embeddings

  • If the tool is breached

    Your CUI is in the blast radius

  • Assessment speed

    Fast

  • What the C3PAO is shown

    A vendor you must now justify as in-scope

Keep it fully manual

Air-gapped folders, spreadsheets, and email chains outside any platform.

  • Where regulated data lives

    Under your control, but scattered

  • What the SaaS can see

    Nothing — because there is no platform

  • If the tool is breached

    No tool to breach; custody is unproven instead

  • Assessment speed

    Slow, and it degrades between assessments

  • What the C3PAO is shown

    Whatever someone reassembled by hand

Athena boundary model

Athena

Regulated data stays in your AWS enclave; only sanitized references cross the gate.

  • Where regulated data lives

    Your AWS Wickr enclave and your retention repository

  • What the SaaS can see

    Identifiers, practice mapping, status, approved conclusions

  • If the tool is breached

    No regulated content to lose — the gate has no return path

  • Assessment speed

    Fast, because the sanitized record is machine-usable

  • What the C3PAO is shown

    A continuously maintained, provenance-backed package

CMMC Level 2 responsibility matrix

Level 2 practices, allocated practice-by-practice across the AWS Wickr enclave, the controlled retention repository, and Athena’s SaaS. Customer configuration and independent validation remain required.

OwnsSharedSupportsOut of boundary

AWS Wickr enclave

0 practices in view

Retention repository

0 practices in view

Athena SaaS

0 practices in view

All

This document describes Athena's use of AWS Wickr as a FedRAMP-authorized component. It does not represent Athena's SaaS or Athena Consulting Group as independently FedRAMP authorized.

A three-layer architecture

Regulated content, retention, and analysis are separate layers with separate authorizations. Nothing is inherited.

Layer 1 — AWS Wickr Premium

FedRAMP Moderate component

the applicable authorized offering is the intended collaboration component for this engagement in the validated AWS region recorded for this engagement. Its FedRAMP authorization scope for this offering and region has not yet been verified against the applicable AWS services-in-scope listing, so no authorization is represented here. Regulated evidence remains in the customer-controlled boundary regardless. End-to-end encrypted messaging, calling, file transfer, and rooms carry the regulated conversation and the CUI itself.

Layer 2 — Controlled retention repository

Customer-configured

The controlled retention repository is a separately configured customer component. It is not represented as FedRAMP authorized merely because AWS Wickr is authorized; its authorization boundary must be determined and documented by the customer. Retention is customer-controlled, keyed with customer-managed keys, and evidenced per service — never inherited.

Layer 3 — Athena assessment-analysis layer

Sanitized metadata only

Athena runs the assessment: objectives, control mapping, findings, POA&Ms, SPRS, and the assessor package. It receives artifact identifiers, sanitized status, and approved conclusions — never regulated content.

What we guarantee in writing

Athena cannot read boundary content

Message bodies, files, and retained-record content are permanently unsupported capabilities. Content reads are refused before any network call and the refusal is audited.

Per-user and customer-managed keys

Vault material is wrapped per user; customers may supply their own key material so the analysis layer never holds a decryptable copy.

Every claim is evidence-backed

Each capability carries a status, the AWS document it came from, and a verification date. Nothing is invoked unless the interface is verified for your deployment.

Fail-closed by default

Mode A (reference-only) ships on. Notification and administrative modes require a recorded Security Officer approval and an in-boundary determination.

Three integration modes, gated in that order

Mode A

Reference-only

Athena records references, statuses, and approvals about the Wickr boundary. It makes no calls to AWS at all. This is the default and the safest posture.

Recording the deployment profile, network reference, and offering or region determination
Recording readiness, approvals, administrator actions, and validation dates
Recording evidence references and sanitized assessment conclusions
Retrieving Wickr message, call, or file content
Retrieving or decrypting retained records
Mode B

Sanitized notification

Athena may send sanitized, allowlisted workflow notifications toward the boundary. Payloads are scrubbed and hashed before delivery, and nothing is read back.

Everything permitted in reference-only mode
Sending allowlisted, scrubbed workflow notifications
Recording the template version, payload hash, and delivery status of each notification
Retrieving Wickr message, call, or file content
Retrieving or decrypting retained records
Mode C

Administrative integration

Athena calls the verified AWS Wickr admin API from server-side code only, to read configuration and health and to evidence readiness. Credentials never reach the browser.

Everything permitted in the lower modes
Reading network, settings, user-directory, security-group, and retention-bot configuration
Reporting configuration drift and health against recorded readiness items
Retrieving Wickr message, call, or file content
Retrieving or decrypting retained records

Verified capability manifest

The same manifest the product enforces at runtime. Only capabilities marked invocable can be called; everything else is documented, not assumed. Last verified 2026-08-05.

CapabilityStatusInvocableSource
Discover Wickr networks
Returns network configuration metadata only. No message or file content is exposed by these operations.
Verified API YesAWS Wickr API Reference
ListNetworks / GetNetwork
Read Wickr network settings
Configuration posture only. Athena reads settings to evidence readiness; writes stay administrator-assisted.
Verified API YesAWS Wickr API Reference
GetNetworkSettings / UpdateNetworkSettings
Read Wickr user directory
Identity metadata only. Used to evidence named-user and device readiness items.
Verified API YesAWS Wickr API Reference
ListUsers / GetUser / ListDevicesForUser
Provision and deprovision Wickr users
Available, but gated behind Security Officer approval because deprovisioning is destructive to an active engagement.
Verified API YesAWS Wickr API Reference
BatchCreateUser / BatchDeleteUser / BatchToggleUserSuspendStatus / BatchReinviteUser
Create and update security groups
Security-group posture, including burn-on-read and federation settings, is readable and writable through the admin API.
Verified API YesAWS Wickr API Reference
ListSecurityGroups / CreateSecurityGroup / UpdateSecurityGroup / ListSecurityGroupUsers
Register OIDC single sign-on
Registration and a dedicated test operation both exist. The client secret must live in AWS Secrets Manager; Athena stores only a reference.
Verified API YesAWS Wickr API Reference
RegisterOidcConfig / RegisterOidcConfigTest / GetOidcInfo
Microsoft Entra ID single sign-on configuration
AWS documents configuring an OIDC provider such as Microsoft Entra ID for named-user sign-on. The identity provider sits outside the AWS authorization boundary and is validated separately; Athena records the registration and its test result, never the client secret.
Verified configuration YesAWS Wickr Administration Guide, single sign-on configuration
Console-guided or OIDC registration
Create and manage the data retention bot
Bot lifecycle is API-manageable. Athena never receives the bot challenge value or any retained content.
Verified API YesAWS Wickr API Reference
CreateDataRetentionBot / GetDataRetentionBot / UpdateDataRetention / DeleteDataRetentionBot
Serverless data retention (Nitro Enclaves and a customer managed key)
Wickr decrypts inside AWS Nitro Enclaves using your customer managed key and writes to an S3 bucket encrypted with that key, read through a provided decryption function. Prefer this over a self-managed container host.
Verified configuration YesAWS Wickr Administration Guide, Data retention modules
Console-guided setup
CloudTrail capture of Wickr API calls
Covers Wickr API and console-originated administrative calls. It does NOT cover in-application client events, so audit claims must be scoped to the control plane.
Verified configuration YesAWS Wickr Administration Guide, Logging with CloudTrail
CloudTrail trail
Read Wickr message or file content from the SaaS
Every message, call, and file is encrypted with a new random key and no party other than the intended recipients, including AWS, can decrypt it. Athena must never attempt to retrieve content.
Unsupported NoAWS Wickr Administration Guide, end-to-end encryption
Deploy a write-only Wickr bot
A data retention bot exists precisely to receive all network content. There is no documented read-restricted bot, so a write-only claim cannot be made.
Unsupported NoAWS Wickr Administration Guide, Data retention modules
Audit in-application Wickr client events
CloudTrail records API activity, not end-user client behaviour inside the Wickr application.
Unsupported NoAWS Wickr Administration Guide, Logging with CloudTrail
Deep link directly into a Wickr room
No documented deep-link scheme. Until AWS confirms one, the product links to the Wickr application generally and names the room in text.
Pending AWS confirmation NoNot documented as of the verification date
FIPS-validated Wickr API endpoint
Regional endpoints follow the admin.wickr.{region}.amazonaws.com pattern. A separate FIPS endpoint is not documented, so no FIPS endpoint claim may be made.
Pending AWS confirmation NoNot documented as of the verification date
AWS Config inspection of internal Wickr settings
AWS Config can record surrounding customer resources such as the retention bucket, key, and roles, but not settings inside the Wickr network.
Unsupported NoNot documented as of the verification date
Authorized offering and region determination
Authorization depends on the specific offering and region the customer deploys. The customer must confirm scope and attach the evidence; Athena records the determination and never asserts it.
Customer-configured NoAWS Services in Scope of the FedRAMP program

What every administrator affirms before activation

  • 01Athena's SaaS is not FedRAMP authorized.
  • 02AWS Wickr's authorization does not extend to Athena's SaaS.
  • 03Regulated evidence will remain within the approved Wickr and retention boundary.
  • 04Only sanitized or customer-approved data may be released into Athena's SaaS.
  • 05Customer configuration and independent validation remain required.
  • 06No regulated processing may begin until the readiness gate is approved.

This document describes Athena's use of AWS Wickr as a FedRAMP-authorized component. It does not represent Athena's SaaS or Athena Consulting Group as independently FedRAMP authorized.

Questions assessors and contracting officers ask

Is Athena Fortify-CMMC FedRAMP authorized?

No. Athena does not represent Fortify-CMMC as FedRAMP authorized. Athena has intentionally architected CUI and regulated evidence out of its SaaS. For regulated engagements, AWS Wickr is used as the authorized secure-collaboration component only after the applicable offering, region, authorization scope, configuration, and supporting AWS services have been validated. CUI and regulated evidence remain within the approved Wickr and customer-controlled retention environment. Athena manages the CMMC workflow through approved metadata, external references, responsibilities, status, and sanitized conclusions. This design gives the customer a clear, defensible system boundary: regulated content stays in the approved environment, while Athena manages the work required to prepare for and sustain CMMC assessment readiness.

Where does CUI actually live?

Sensitive evidence lives in the approved Wickr and retention boundary. Athena’s SaaS manages sanitized workflow, status, control mapping, and approved conclusions.

Can Athena decrypt anything inside the boundary?

No. Capabilities that would read message bodies, file content, or retained-record exports are marked unsupported in the capability manifest and are refused locally, before any request is formed. The refusal is written to the boundary audit log.

Does Wickr's authorization cover the whole stack?

The controlled retention repository is a separately configured customer component. It is not represented as FedRAMP authorized merely because AWS Wickr is authorized; its authorization boundary must be determined and documented by the customer.

How is a new integration approved?

Verify-before-build. An interface is only invoked once the AWS documentation for it is recorded, the per-tenant capability manifest is verified for the offering and region, and the Security Officer approval is on file.