HoundShield is a local-only AI compliance firewall. It sits between your team and tools like ChatGPT, Copilot, and Claude, scans every prompt on your own hardware in under 10 milliseconds, blocks CUI, PHI, and PII before anything leaves your network, and writes tamper-evident audit evidence mapped to NIST 800-171.

How does HoundShield stop employees pasting CUI into ChatGPT?

HoundShield works as an OpenAI-compatible proxy: you change one base URL and every AI request passes through 16 local detection engines before it leaves your perimeter. Prompts containing CUI markings, CAGE codes, contract numbers, or clearance terms are blocked instantly and logged to a SHA-256 hash-chained audit trail.

Does HoundShield send my prompts to the cloud?

No. In self-hosted Docker mode, scanning happens entirely on your infrastructure and prompt content never leaves your network — unlike cloud DLP tools, which must receive your data to inspect it. Only the hosted trial runs in our cloud, and it is for demos and non-CUI evaluation only.

Who is HoundShield for?

HoundShield is built for regulated teams that use AI: defense contractors preparing for CMMC Level 2, healthcare organizations protecting PHI under HIPAA, and law firms guarding privileged communications. If your staff use ChatGPT, Copilot, or Claude and an auditor will ever ask for evidence, HoundShield is for you.

How quickly can we get started with HoundShield?

Most teams deploy in under 15 minutes: run the Docker gateway, point your AI tools' base URL at it, and verify a test prompt in the audit log. The free tier needs no credit card, and the $499 assessment report delivers signed evidence within days of intake.

Local-only · CMMC Level 2 · HIPAA · SOC 2

Stop your team from leaking CUI to ChatGPT.

HoundShield intercepts every AI prompt before it leaves your network. 16 detection engines. Sub-10ms latency. CMMC Level 2, HIPAA & SOC 2 — enforced simultaneously, scanned on your hardware.

Start free — no card required See how it works

One URL change Local-only Free to start C3PAO-ready

HoundShield

Live demo

Live AI Monitor

Scans 24h

142,690

▲ ~46/min

Blocked

2,218

▲ CUI · PII · secrets

SPRS

▲ NIST 800-171

Quarantine

awaiting review

Gateway throughput prompts/sec

Prompts flowing through the local gateway, scanned in <10ms each.

Detection mixtoday

What the blocks were — CUI, secrets, PII and PHI.

2,218blocked

CUI 34%

Secrets 24%

PII 27%

PHI 15%

Detections by engine live

Which of the 16 detection engines are firing right now.

CUI

Secrets

PII

PHI

Source/IP

CAGE

Where prompts go24h

The AI tools your team uses — every prompt scanned locally first.

ChatGPT

46%

Copilot

31%

Claude

18%

Other

CMMC Level 2 coverage — DoD supplier score, live from the assessmentSPRS 84 · 88%

Live prompt scanson-device

BLOCKEDCUI · CAGE 1ABC2 + contract #7ms

PASSEDclean · weekly status email11ms

BLOCKEDSecret · sk-live-…a91 API key5ms

QUARPHI · patient MRN 44719028ms

PASSEDclean · summarize public RFP12ms

Detection engines

CUI · PHI · PII · IP

~80,000

Contractors at risk

Need CMMC Level 2

110

NIST 800-171 controls

Mapped & SPRS-scored

<10ms

Scan latency

Median, fully local

CUI-safe = Mode B (Docker on your infrastructure). The hosted trial runs on Vercel and is not FedRAMP-authorized — use it for non-CUI evaluation only.

The asymmetric advantage

Cloud DLP scans your CUI in their cloud. That's the spill.

Every cloud-based AI DLP tool has to receive your data to inspect it. For a DoD contractor, that transmission is itself a DFARS 7012 CUI exposure. HoundShield is the only one that never sees your data.

Nightfall & Strac

Cloud DLP. To scan a prompt they must transmit your CUI to their servers — the exact exposure CMMC L2 forbids. Architecturally disqualified for the DIB.

Microsoft Purview

M365-only. No API proxy. Your team's ChatGPT, Claude, Cursor and Copilot traffic outside Microsoft's walls goes completely unmonitored.

HoundShield

Local-only. Detection runs on yourhardware. Nothing reaches our servers — ever. The moat cloud vendors can't match without a full rebuild.

One platform

Everything you need for CMMC Level 2

Map, assess and close gaps across all 110 NIST 800-171 controls — with AI-driven remediation and C3PAO-ready evidence.

110 controls

CMMC Self-Assessment

Guided questionnaires across all 110 NIST SP 800-171 controls. Your SPRS score updates live as you complete each practice.

Prioritized

AI-Powered Gap Analysis

Brain AI flags unmet controls and generates a remediation roadmap ranked by risk severity and cost — on-device, your key.

1-click

SSP & POA&M Export

Auto-generate your System Security Plan and Plan of Action & Milestones as C3PAO-ready PDFs with SHA-256 signed evidence.

AI Prompt Interception

Every LLM request inspected before it leaves the perimeter. Works with ChatGPT, Copilot, Claude, Gemini — all at once.

16 Detection Engines

CUI, PII, IP, PHI, secrets, CAGE codes, contract numbers and clearance markers — flagged, blocked or quarantined.

Live Threat Dashboard

Real-time blocked prompts, risk scores and compliance posture for every employee, in one command center.

Start here · one-time report

CMMC AI Risk Assessment Report

Your staff are pasting sensitive data into ChatGPT and Copilot right now — with no audit trail. We scan 14 days of your real AI traffic locally and hand you the signed evidence your assessor asks for. No subscription, no procurement cycle.

14 days of real AI-traffic scanning across 16 local detection engines (CUI · PHI · PII · ITAR)
Every prompt event scored against NIST 800-171 Rev 2 with your SPRS impact
SHA-256 hash-chained audit trail — tamper-evident, assessor-defensible
C3PAO-ready PDF plus a 30-minute readout of findings and next steps

Mode B (Docker) — data never leaves your network Delivered within days of intake No signup required

$499

one-time · per organization

Talk to us first →

Fixed price. Yours to keep. See a sample report (PDF)

FAQ

Questions teams ask before deploying

Still have questions? Talk to a compliance engineer — we respond within 4 business hours.

Ready to protect your CUI?

No credit card required. Assess all 110 controls and see your SPRS score in under 30 minutes.

Get started free

CMMC AI Risk Assessment Report

14 days of real AI-traffic scanning across 16 local detection engines (CUI · PHI · PII · ITAR)

Every prompt event scored against NIST 800-171 Rev 2 with your SPRS impact

SHA-256 hash-chained audit trail — tamper-evident, assessor-defensible

C3PAO-ready PDF plus a 30-minute readout of findings and next steps

Mode B (Docker) — data never leaves your network Delivered within days of intake No signup required