NIST Alignment

FEDERAL · NIST ALIGNMENT

APIs / Identity / AI agents

Aligned with NIST,
from web services
to AI agents.

NIST guidance has moved from securing SOAP web services in 2007 to protecting cloud-native APIs and governing AI agents today. QS Platform 10 maps to that whole arc: FS Sentry for API protection and zero trust, QS DevX for inventory and governance, and QS Cortex for AI risk management.

This page replaces our 2017 white paper on SP 800-95 with a current view of the NIST publications that matter for API and AI security, and how the platform supports each one.

No registration required. Updated November 2026.
QS PLATFORM 10 · NIST MAPNIST
QSPlatform 10

FS Sentry

SP 800-228 · 800-95

QS DevX

Inventory · governance

QS Cortex

AI RMF · AI 600-1

SP 800-207 zero trust · SP 800-63-4 identity · SP 800-53 controls
APIs
SP 800-228
Identity
SP 800-63-4 · 800-157
AI
AI RMF · IR 8596
One platform for the APIs and the AI that uses them.
FROM SP 800-95
TO AGENTIC AI.
API protection

SP 800-228 pre-runtime and runtime controls

Zero trust and identity

SP 800-207, 800-63-4 and derived PIV

AI risk

AI RMF, GenAI Profile and agent security work

01

TWO DECADES OF NIST GUIDANCE

The guidance moved
with the architecture.

Each wave of architecture brought new NIST guidance: web services, microservices, zero trust, cloud-native APIs, and now generative and agentic AI. The constant is the API, which is where AI agents act.

  1. 2007SP 800-95Guide to Secure Web Services: SOAP and XML security
  2. 2019SP 800-204Security strategies for microservices-based applications
  3. 2020SP 800-207Zero Trust Architecture and the policy enforcement point
  4. 2023AI RMF 1.0AI Risk Management Framework: govern, map, measure, manage
  5. 2024AI 600-1 · SP 800-218AGenerative AI Profile and secure development for GenAI
  6. 2025SP 800-228 · SP 800-63-4API protection for cloud-native systems and digital identity
  7. 2025AI 100-2 E2025 · IR 8596Adversarial ML taxonomy and the draft Cyber AI Profile
  8. 2026CAISI · COSIS · FIPS 140-3Agent standards, SP 800-53 AI overlays, FIPS 140-2 sunset
02

API GUIDANCE · NIST SP 800-228

API protection for cloud-native systems,
pre-runtime and runtime.

Published in June 2025 and updated in March 2026, SP 800-228 is NIST’s current guidance for API protection. It recommends basic and advanced controls across the API lifecycle and describes enforcement patterns including API gateways, service meshes, sidecars and egress gateways. The control areas below summarize its themes and how QS Platform 10 supports them.

Control areaWhat the guidance emphasizesHow QS Platform 10 supports it
API inventory and classificationKnow every API, its owner, exposure and sensitivity, including internal and third-party APIs.QS DevX catalog of APIs and gatewaysQS DevX
Authentication and authorizationVerify every caller and authorize every request, with consistent identity across credential types.Any-to-any identity translation, RBAC, ABAC and CBACFS Sentry
Schema and input validationValidate requests and responses against API definitions and reject malicious input.Schema validation, hardened parser and threat inspectionFS Sentry
Resource protectionLimit rate, size and consumption to prevent abuse and denial of service.Rate, size and SLA controls per consumer and agentFS Sentry
Sensitive data protectionPrevent sensitive data from leaking through API responses and logs.Encryption, redaction and data leakage preventionFS Sentry
Egress controlControl and inspect outbound calls to third-party APIs and services.Protocol break and controlled outbound routingFS Sentry
Monitoring and loggingLog and monitor API activity to detect misuse and support response.Full-context, bi-directional transaction logsFS Sentry · QS DevX
Lifecycle and DevSecOpsApply controls from design and testing through deployment and runtime.API testing, Git policy promotion and hot-swap deploymentQS DevX · FS Sentry

Summary based on NIST SP 800-228. Organizations should consult the publication for the full set of recommendations.

Gateway as identity translator. SP 800-228 recognizes the API gateway as an enforcement point that can normalize diverse credentials, a role FS Sentry has filled for two decades.

API security & management
03

STILL RELEVANT · NIST SP 800-95

The web services threats
haven’t gone away.

SP 800-95 (2007) remains NIST’s guide to securing SOAP and XML web services, which still carry mission-critical government traffic. FS Sentry has addressed each threat it lists, drawn from WS-I, since the beginning.

Message alterationSchema validation and digital signature verification preserve message integrity.
Loss of confidentialityData-level and transport encryption, hashing and pattern-based redaction.
Falsified messagesMutual TLS authentication and replay detection for message-based security.
Man in the middleTLS termination and initiation with X.509 path validation, revocation checking and host verification.
Principal spoofingSignature verification bound to designated digest locations in the message.
Forged claimsCredentials authenticated and authorized against enterprise identity systems, with token correlation.
Replay of messagesReplay detection for WS-Security tokens and signed messages.
Replay of message partsIntegrity checks that prevent valid signatures being reused elsewhere in a message.
Denial of serviceRate- and size-based enforcement, with intrusion rules that throttle or block offenders.

Threat list from NIST SP 800-95, Section 2.6. Our 2017 white paper covers SP 800-95 in full and remains available as an archive.

04

ZERO TRUST, IDENTITY AND CONTROLS

Verify every caller,
human, service or agent.

API protection rests on NIST’s zero trust, identity and control guidance. These publications define what a policy enforcement point does, how identities are proven and federated, and the controls auditors assess.

SP 800-207 · 800-207A

Zero Trust Architecture

Every subject reaches resources through a policy decision point and a policy enforcement point; 800-207A applies this to cloud-native, multi-cloud environments.

QS Platform 10: FS Sentry is a cyber-hardened PEP for people, services and AI agents.

Learn more
SP 800-63-4

Digital Identity Guidelines

Requirements for identity proofing, authentication and federation, updated in 2025.

QS Platform 10: SSO, federation, MFA with built-in TOTP, and continuous authentication.

Learn more
SP 800-157

Derived PIV Credentials

PKI credentials derived from a PIV card for use on mobile devices.

QS Platform 10: FS Sentry validates derived PIV credentials for cloud and on-premises APIs.

Learn more
SP 800-53 Rev. 5

Security and Privacy Controls

The control catalog behind FedRAMP and federal assessments, including access control, identification and authentication, audit and system protection families.

QS Platform 10: Gateway enforcement and full-context logs that support evidence for these control families.

05

CRYPTOGRAPHY · FIPS 140

Validated as a whole product,
not just a card.

Many products point to a FIPS-certified hardware security module and call themselves FIPS compliant. That certifies the card that stores keys. It says nothing about the code that parses messages, makes policy decisions and enforces security on every request. FS Sentry took the harder path.

FS SENTRY

The entire product as the cryptographic boundary.

FS Sentry was validated to FIPS 140-2 Level II with the whole product inside the cryptographic boundary. Every part of the codebase, including the parsers, policy engine and enforcement logic, went through the evaluation, not just the module that holds keys.

That rigor shaped the engineering practices behind the product, and FS Sentry hardware also integrates an HSM card certified to FIPS 140-2 Level III for key storage.

HSM CARD ONLY

  • Certifies key storage on the card
  • Parsing, policy and enforcement code never evaluated
  • FIPS claim stops at the card boundary

WHOLE-PRODUCT BOUNDARY

  • Entire codebase evaluated as one boundary
  • Policy and enforcement inside the evaluation
  • Level III HSM card for key storage as well
The FIPS 140-3 transition. NIST’s Cryptographic Module Validation Program stopped accepting FIPS 140-2 submissions in 2022, and on September 21, 2026 all remaining FIPS 140-2 certificates, including FS Sentry’s, moved to the CMVP Historical list. Existing systems may continue to use historical modules; new federal procurements generally look for active FIPS 140-3 validation. Talk to us about the cryptographic requirements of your program.

Also certified: Common Criteria NDPP and DoD JITC PKI certification for the gateway.

Federal API security

06 · AI GUIDANCE

NIST’s AI guidance,
mapped to the platform.

NIST’s AI work now spans risk management, generative AI, adversarial machine learning, secure development, a draft Cyber AI Profile, SP 800-53 overlays for AI systems and an initiative focused on AI agents. Much of it is still evolving; the direction is clear: identity for agents, least-privilege access and evidence of every action.

AI RMF 1.0
Govern, map, measure and manage AI risks across the lifecycle.Policies, entitlements and traces across QS Cortex and QS DevX support governance and measurement.
NIST AI 600-1
Generative AI Profile: risks such as information security, data privacy, confabulation and value chain.Guardrails, obfuscation, grounding in governed data and approved models in QS Cortex.
NIST AI 100-2 E2025
Adversarial machine learning taxonomy, including prompt injection and poisoning.Input guardrails, tool allowlists and deny-by-default enforcement that limit what a manipulated agent can do.
SP 800-218A
Secure software development practices for generative AI and foundation models.Vetted packages in a sandboxed runtime and tested, generated tools in QS DevX.
NIST IR 8596 (draft)
Cyber AI Profile: securing AI systems, AI-enabled defense and thwarting AI-enabled attacks, on CSF 2.0.One platform securing AI systems and the APIs they touch, with monitoring and logging throughout.
SP 800-53 overlays (COSIS)
Control overlays for AI systems, including single-agent and multi-agent use cases, in development.Agent identity, least-privilege tool access and audit trails that map to SP 800-53 control families.
CAISI AI Agent Standards
Agent security, identity and authorization, and interoperability research launched in 2026.Agents act with delegated user identity; every action is authorized by FS Sentry and traced in QS DevX.

Draft and in-development NIST publications may change. COSIS was previously named COSAiS (Control Overlays for Securing AI Systems).

07

RESOURCES

Read the guidance
at the source.

Links to the NIST publications referenced on this page, plus our archived white paper.

NIST does not certify or endorse commercial products. Alignment describes how platform capabilities support NIST recommendations; compliance depends on each organization’s implementation and assessment.

THE DETAILS

Questions,
answered.

NIST guidance for APIs, identity, cryptography and AI.

Sovereign AI for government
Which NIST publication covers API security today?

NIST SP 800-228, Guidelines for API Protection for Cloud-Native Systems, published in June 2025 and updated in March 2026. SP 800-95 (2007) still applies to SOAP and XML web services.

Does NIST certify API gateways or AI platforms?

No. NIST publishes guidance and standards. Cryptographic modules are validated through NIST’s CMVP under FIPS 140-3, and organizations demonstrate alignment through their own implementation and assessments.

Is FS Sentry FIPS validated?

FS Sentry was validated to FIPS 140-2 Level II with the entire product as the cryptographic boundary, so the full codebase, not just a key-storage card, was evaluated. Its hardware also integrates a FIPS 140-2 Level III certified HSM card. Like all FIPS 140-2 certificates, the validation moved to the CMVP Historical list on September 21, 2026.

What happened to FIPS 140-2?

All remaining FIPS 140-2 certificates moved to the CMVP Historical list on September 21, 2026. Existing systems may continue to use historical modules, and FIPS 140-3 is the active standard for new validations.

What NIST guidance applies to AI agents?

The AI RMF and Generative AI Profile, the adversarial ML taxonomy (AI 100-2), the draft Cyber AI Profile (IR 8596), SP 800-53 overlays for AI systems in development, and CAISI’s AI Agent Standards Initiative launched in 2026.

How does QS Platform 10 support NIST zero trust guidance?

FS Sentry acts as the policy enforcement point described in SP 800-207, authorizing every call from people, services and AI agents, while QS DevX entitlements inform policy and full-context logs provide evidence.

Can we still get the SP 800-95 white paper?

Yes. It remains available as an archive, without changes, for teams that need the original SP 800-95 mapping.

NIST ALIGNMENT · QS PLATFORM 10

Map your NIST requirements
to one platform.

Our federal team can walk through SP 800-228, zero trust, identity and AI guidance against your program’s requirements.