Trust centre

Trust is verifiable.

A lock doesn't ask for trust; it shows you the ledger.

How the gateway works
ISO 27001 certified
audit chain · verifysealed
last 4 entriessha-256 linked
019f2a·c41dchat.completion
02c41d·7b03mcp.tool_call
037b03·e18fresidency.block
04e18f·40a6embeddings
chain intact ✓tamper-evident

Nothing sails out until your policy opens the gate. EU by default: enforced, not promised.

Every request meets your policy before it meets a model; if nothing qualifies the gate stays shut: blocked, audited, no dispatch.

  • Sovereign EU

    Mistral and Scaleway are French companies on French infrastructure, outside any non-EU government access regime.

  • Direct US or China

    US and Chinese providers route only when your policy allows, flagged by jurisdiction and CLOUD Act.

  • Your endpoints, your iron

    Bring a private or self-hosted model, or run the whole data plane yourself with Sluis Edge.

residency.map · egress: your policyenforced
POLICYAmsterdam · NLParis · FRFrankfurt DE · US-ownedUS East · allowedCN · not in policy
EU-owned · sovereign (default)EU region · US-owned, flagged (CLOUD Act)allowed by policyoutside policy · blocked

Erase the content, keep the proof.

An erasure request that destroys your evidence isn't compliance, it's a liability: Sluis removes the content and keeps the proof.

compliance erase --tenant <id>chain still verifies
Purged: right to erasure (GDPR Art. 17)
  • Request & response content: the prompts and completions held under opt-in retention.
  • Cache entries: exact and semantic cache rows for the tenant.
  • Provider credentials: stored upstream API keys, decrypted nowhere.
  • Content references: every content_ref pointer is nulled.
Retained: immutable proof
  • The audit metadata chain: audit_log + chain_heads stay intact, so verify-chain still passes.
  • Hashes, not content: the SHA-256 links remain; the bodies they covered are gone.
  • Routing decisions: which region and provider served each call.
  • Token & cost metadata: the real-money micro-euro ledger that backs your budgets.

Retention TTL purge

A scheduled, cron-able purge deletes only the content and cache rows past their TTL.

Content retention is opt-in

By default Sluis records metadata, not the message; storing prompt and response bodies is opt-in.

AEAD-encrypted at rest

Retained content and the entire response cache (exact + semantic) are sealed with authenticated encryption.

Strict per-tenant isolation

Keys, policies, audit chains and cache are scoped to one tenant, enforced on the hot path.

Hand the auditor the chain. They verify it without trusting a word we say.

Every passage is hash-chained to the last, so altering one field breaks every later link; export as JSON Lines and re-verify offline.

audit.chain · last 4 entrieschain verified
14:22:07.118Z#4f9c2a e1b7d9 · scaleway-fr · PHI · maskedin-region
14:22:05.904Z#e1b7d9 7a3f10 · mistral-fr · PIIin-region
14:22:04.661Z#7a3f10 0c55ab · vertex-eu · generalfallback
14:22:02.330Z#0c55ab genesis · scaleway-fr · generalin-region

$ sluis audit verify-chain --tenant <id> → ok · 1,284,901 entries · genesis intact

Provider exposure, region and ownership, stated plainly.

Representative routes show why hosting region and legal ownership are separate controls under laws like the CLOUD Act.

ProviderHosting regionLegal ownerExposure
MistralParis · FREU (France)sovereign
ScalewayParis · FREU (France)sovereign
Google VertexEU multiregionUS (Google)CLOUD Act
AWS BedrockEU regionUS (Amazon)CLOUD Act
Azure OpenAIEU regionUS (Microsoft)CLOUD Act
OpenAIUnited StatesUS (OpenAI)opt-in
AnthropicUnited StatesUS (Anthropic)opt-in
xAI (Grok)United StatesUS (xAI)opt-in
DeepSeekChinaCN (DeepSeek)opt-in
Alibaba QwenChinaCN (Alibaba)opt-in
Moonshot AIChinaCN (Moonshot)opt-in
Zhipu AIChinaCN (Zhipu)opt-in
Per-tenant disclosure is exportable on demand: compliance subprocessors --tenant <id> returns only the providers your keys and policy actually enable.

Sensitive data is caught at the first gate, before it reaches a model.

60 deterministic detectors scan every request for PII, PHI and secrets, built for regulated data everywhere: from the US Social Security number to a national-ID pack that checksum-validates 12 EU countries.

The same engine anonymizes whole documents: docx, pdf, images, and text go in, the same document comes out with PII as «MERGE_TAG»s and image regions blurred, rendered and OCR'd inside the gateway so nothing ever reaches a provider. Redacted PDFs keep an invisible searchable layer of the anonymized text, and async jobs deliver results through time-limited signed URLs.

PII detectors
emailE.164 phoneUS SSNIBAN · mod-97NL BSN · 11-proefPL PESELIT codice fiscalecredit card · Luhn
Secret detectors
API keysAWS access keysPEM private keys
Entity detection

Names defeat pattern matching, so detection is five layers you switch on deliberately: context heuristics, email correlation, a tenant name directory, a shipped name dictionary compiled from government open data, and a recognition model that runs inside the deployment, so text never leaves your perimeter. Two selectable model tiers trade latency for catch rate: fast (spaCy) or deep (GLiNER2-PII).

We publish what each tier actually catches: measured accuracy per model and language, method and caveats included, in the docs benchmark table.

Security scanning

Opt-in prompt-injection and jailbreak detection runs at the gate in log or block mode, with an org-configurable threshold; hits are sealed in the audit trail and the scan fails open. Opt-in key-behaviour anomaly detection runs as a background job with per-key baselines and an isolation-forest layer; alerts are explainable, never a bare score, in the console Security tab and optionally by email.

Built with Llama. The prompt-injection scan model is Llama Prompt Guard 2 86M, used under the Llama 4 Community License.

dlp.mode · per tenantrequest-side
mask
Rewrite the matched spans in place, then forward the redacted request upstream.
mode
block
Refuse the call at the gate (422, nothing leaves) when a class is not permitted.
mode
allow-log
Pass the request through, but note the detection in the audit trail for later review.
mode
pseudonymize
Swap each value for a reversible token before dispatch; the model never sees the real one.
mode
Reversible pseudonymization

Work with personal data and secrets. The model never sees them.

PII and secrets become stable tokens like «EMAIL_1» before the prompt leaves your network, and the map restoring them never touches disk.

your apprequestSluis gatejohn@acme.com«EMAIL_1»«EMAIL_1»masks PIImodel«EMAIL_1»«EMAIL_1»token onlySluis gate«EMAIL_1»john@acme.comjohn@acme.comrestoresresponseyour appyour app · requestSluis masksjohn@acme.com«EMAIL_1»«EMAIL_1»PII stays insidemodel«EMAIL_1»«EMAIL_1»token onlySluis restores«EMAIL_1»john@acme.comjohn@acme.cominside the streamyour app · restored

Everything your reviewers ask for, ready to hand over.

The paperwork that usually takes three weeks: request any of it directly, no sales call needed to read a DPA.

Security contact
Reach the security team at security@sluis.ai. We operate coordinated, responsible disclosure: report a vulnerability privately and we will acknowledge it, keep you posted, and credit you once it's fixed.
Report an issue

Bring your auditors. We'll bring the chain.

Residency enforced on every request, content you can erase without losing the proof, and a ledger anyone can check offline. Come test them.

Regulated buyer? Ask about Sluis Edge →

Read the docs