Get Started Overview Proofs Healthcare Financial Services Legal Education E-Commerce Integrations Startups Enterprise

AI tools leak your data. Shield stops it — before anything leaves your company.

Every time your team uses ChatGPT, Claude, Copilot, or any AI tool, sensitive information can slip into the prompt. Customer records, financial data, patient names, legal documents — sent to external servers you don't control. Shield installs on your machines and catches it all automatically.

No cloud service. No subscription. One install, every AI tool on that machine is protected. Priced flat — $10K to $45K, not per seat.

By Purfect AI · Runs on your network

Shield helps teams deploy AI safely, with clear controls, real-time protection, and AI-native discoverability.

Redact your secrets before they reach the model.

Purfect Shield is a transparent redaction gateway that sits in front of every LLM tool your team uses. The secrets and PII your filter packs match are replaced with placeholders outbound, then rehydrated inbound — so the model provider never receives those matched values in cleartext.

No hosted service. No license server. No subscription. You own the source.

$10K/$25K/$45K, two PO line items, published and flat — no per-seat or per-token meter.

Purfect Shield — Product Demo

Two layers. Two fail-safes. A developer never knows redaction happened — they write the same code, paste the same context, hit the same enter key.

The leak surface

Your team uses AI every day — ChatGPT for drafting, Copilot for coding, Claude for analysis, Notion AI for notes. And every day someone accidentally includes customer data, financials, or internal documents in a prompt. The AI provider stores it. Your compliance team can't see it. Your security team can't stop it.

Cloud credentials

AWS_ACCESS_KEY_ID, GCP service-account JSON

Source-control tokens

ghp_..., GitLab PATs, gho_...

API keys

sk-ant-..., sk-..., internal service tokens

Environment secrets

.env contents read into context by a tool

Database credentials

Postgres / Mongo connection strings

JWTs

eyJ... tokens in headers, args, fixture files

Domain PII

SSNs, patient names, DOBs, card PANs

Tool exfiltration

Tool trying to read ~/.ssh/, ~/.aws/, /etc/shadow

Asking your team to manually sanitize every prompt is asking them to stop shipping.

The Engagement Week

Foundation deploys in 5 days. Not months, not quarters — five working days from kickoff to production-ready.

Day 1

Threat model workshop

We sit with your team, map your actual attack surface — your prompts, your tools, your data. Select your domain filter pack.

Day 2

Protection active

Shield is running on your systems. Every AI tool on protected machines is now inspected automatically.

Day 3

Tools connected

All your team's AI tools are wired through Shield. ChatGPT, Claude, Copilot, Cursor — one setup covers them all.

Day 4

Verified and tested

We run 30+ real-world scenarios against your environment. Every test passes, every filter works, you see the proof.

Day 5

Your team, trained and running

Runbook delivered. Source code is yours. Your team knows how to add new tools, new filters, and new users. No vendor lock-in — you own it.

How we engage

We work with your team — whether that's your CTO, compliance officer, IT director, or department head. If your organization is adopting AI and worried about what's leaking, we deploy with whoever owns that risk. Engineering handles the technical install; leadership gets the compliance evidence. Everyone stays in their lane.

What the model saw

[REDACTED:AWS_KEY_001]
[REDACTED:PHI_NAME_001]
[REDACTED:JWT_001]
...sanitized payload

Redacted tokens. Sanitized payloads. Zero PHI.

What your developer saw

AKIA2X9...REAL_KEY
Sarah Johnson
eyJhbGci...full_token
...full response

Real values. Full responses. Nothing hidden from the developer.

Belt and suspenders — plugin catches obvious leaks at the agent-loop boundary, gateway scrubs the wire. If one layer misses, the other catches.

Three Tiers, One Standard

All tiers include source delivery. No subscription, no license server, no kill switch. Upgrade between tiers at any time — prior engagement credited.

🚦

Foundation

$10,000 · 1 day

One LLM provider. One day. You own the source.

Gateway deployed on your infra, belt-and-suspenders plugin installed, threat model workshop, one domain filter pack, 30+ eval cases, runbook, and 30-day support. Source delivered at handoff — yours forever.

→Gateway (localhost:9090) deployed on your infra

→Plugin auto-installed for 1 LLM CLI (Codex, Cursor, DeepSeek, Ollama, Windsurf)

→Threat model workshop — Day 1, with your team

→Default filter packs: generic secrets, AWS credentials, GitHub tokens, common PII

→1 domain filter pack — healthcare, finserv, legal, education, or ecommerce

→30+ client-specific eval cases, all green

→Runbook + full Go source delivery

→30-day email support

Contact Sales →

Add-ons

Available on any tier. Semantic Sentiment Filter is included free at Enterprise.

Additional LLM provider$2,000

Additional filter pack$3,000

Semantic Sentiment Filter (on-prem ONNX, 5-axis)$5,000

S3 Federation upgrade (Compliance → Federated)$8,000

CI eval gate — single pipeline$3,000

CI eval gate — multi-pipeline / multi-repo$5,000

Retainer (async + office hours)$5,000/mo

Annual refresh (source + filter pack updates)$1,000/yr

🔒

Your data never touches our infrastructure

The gateway runs on your network. The audit log writes to your S3 with your encryption keys under your retention policy. There is no Purfect Shield cloud. There is no telemetry. Audit teams query their existing pipeline.

SOC 2 evidence readyHIPAA audit trailYour S3 + your keysAppend-only logNo license serverSource code delivered

Intake Portal

Let's Build.

Submit your technical details and we will formulate a production scope, architectural dependencies, and exact model selection profiles.

48-Hour Response SLA

Every request is routed directly to a principal systems engineer.

Zero Cloud Risk

All contracts guarantee strict IP ownership and security boundary isolation.

Frequently Asked Questions

Everything you need to know about Purfect Shield

What happens if Purfect Labs goes out of business?

You own the source. The perpetual license means you self-maintain forever. The Go gateway compiles with a single command. No license server, no kill switch, no call-home.

Do you have SOC 2?

SOC 2 is not applicable by architecture — we hold no keys, run no servers, process no data. Shield runs entirely on your infrastructure. See our architecture letter for details.

How is this different from LLM Guard?

Structural, not chronological. Shield is a Go egress proxy (one env var); LLM Guard is a Python library (code change per integration). Shield sits at the network boundary — no per-application integration required.

What AI providers do you support?

Any OpenAI-compatible endpoint. Anthropic, OpenAI, DeepSeek, local models via Ollama/LM Studio. Shield is provider-agnostic — if the API shape matches, Shield scrubs it.

How much does it cost?

$10K/$25K/$45K — published, flat, no per-seat or per-token meter. Two PO line items. Upgrade between tiers at any time with prior engagement credited. No subscription, no renewal fees, no hidden costs.

Can we try before buying?

Yes — we run a live demo on YOUR LLM flow, in YOUR environment, with YOUR prompt shapes. It's a working gateway, not a slide deck. You see real redaction on real prompts before committing.