Policy before inference. Evidence after.

Set organisation-scoped routing, optional usage limits, and review what was used and what changed. The same rules apply to every request, whichever consumer sent it.

D×E routing policy

Declare request sensitivity (D-class), assign provider environment classes (E-class), and use the routing matrix to block incompatible requests before inference. API keys enforce sensitivity defaults and floors.

Explore data routing →

Usage limits

Optionally cap costs and tokens per member or dedicated API key and per provider. Without a cap, no optional usage limit applies; routing and technical limits still apply. Steinkauz AI does not include inference credit.

Explore Usage limits →

Usage and audit visibility

Inspect inference activity across every consumer, review configuration audit events for policy and provider changes, and trace administrative actions.

Explore audit visibility →

Routing

Data routing policy

Choose which environments may process each declared sensitivity level. Incompatible API requests are blocked before inference.

Interactive demo. No data is saved.

Docs

API sensitivity

The key default is used without a request declaration. The declaration must meet the key floor; this demo fixes the floor at D0. Classify the entire request, including any transient file input, in your application.

Used when the request has no sensitivity header.

Sent using X-Steinkauz-Data-Sensitivity; applies to the entire request.

Provider

Each provider has a provider environment class (E-class). Pick one to simulate a request.

Simulate a request

Compare using the key default with an explicit request declaration.

Effective sensitivity: D0: Public

D×E routing matrix

Configure which provider environment classes may handle each data sensitivity level. Toggle cells to allow or block combinations.

AllowedBlocked
Data sensitivityE0E1E2E3E4E5
D0
D1
D2
D3
D4
This request would be allowed at D0: Public.

API keys also enforce a minimum data class and execution environment. A declaration cannot go below the key floor. Labels are customer assertions; the Platform does not classify file contents.

Usage limits

Optional ceilings, not included credit

Cap cost and/or tokens per member or dedicated API key, per provider. Empty fields add no optional cap; other limits still apply. Remaining 0 blocks further requests on that provider.

  • Caps are optional and per member or dedicated API key × provider, not a single global limit, and not included inference
  • Cost limits and token limits can be set independently; skip a field and that track stays uncapped
  • Inherit keys use their creator’s usage limits; dedicated keys have their own ceilings
  • Existing tools, custom software, through the API all share the same rules. Limits reset each UTC calendar month
Usage limits
September 2026

Cap cost and/or tokens per member or dedicated API key, per provider. Empty fields add no optional cap; other limits still apply. Remaining 0 blocks further requests on that provider.

Anthropic

Maya Chen

Member

On track
Cost ($)$58.00 / $200.00
Tokens in1.2M / 3.0M

Tom Weber

Member

Low remaining
Cost ($)$68.00 / $80.00
Tokens in410K

Uncapped

billing-bot

Dedicated API key

Exhausted
Cost ($)$50.00 / $50.00
Tokens in820K / 1.0M

Azure OpenAI

Maya Chen

Member

No cap set
Cost ($)$12.40

Uncapped

Tokens in240K

Uncapped

Preview only. Owners and admins configure usage limits in the app under Usage limits. Limits reset each UTC calendar month.

Audit

Audit built in

Two complementary surfaces: AI inference activity and configuration changes.

AI inference activity

  • Request detail: read-only input/output log, request metadata, routing decision, usage and safe file evidence
  • Dedicated activity page for searchable requests across API with API-key attribution
  • Model-call timing and cost, plus JSON evidence downloads subject to content retention and role permissions
  • Unified usage and cost visibility for reconciliation and review
Activity details

Governed API requests have searchable policy, attribution and usage evidence.

API
SuccessUser messageD2: ConfidentialE1: Public managed
Modelclaude-opus-4-6
ProviderAnthropic
Tokens1,650
Cost$0.00234
Duration2.3s
Steps3
Request IDreq_8f2a…c91d
Policy decision
Allowedpde_8f2a…c91d

Open Activity to inspect inputs, outputs, request metadata, the policy decision and usage in one detail, subject to retention and role permissions.

Search and filter requests by model, member, API key, connection, decision and processing status. Inspect inputs, outputs, timing, cost and policy evidence with the appropriate permissions.

Configuration audit log

Every meaningful policy or provider change leaves a trace for review.

Routing policyToday, 09:14

Routing matrix changed: D3: Restricted → E1: Public managed, allowed → blocked.

API accessYesterday, 16:42

API key created with default sensitivity D2: Confidential and minimum sensitivity D1: Internal.

Usage limitsYesterday, 11:08

Optional cost cap set for Alex on Azure OpenAI.

ProviderMon, 08:55

Azure OpenAI provider environment class updated to E3: Private cloud.

Routing, connection, usage limit, API access and membership changes are available in Change history.

Configuration audit

  • Routing matrix changes with field-level before/after comparison
  • Usage limit audit for cap edits on members and API keys
  • Provider audit for creation, updates, deletion, and environment-class changes
  • Administrative audit for membership and billing-sensitive organization changes

Govern AI without losing agility

Set policy once. Apply it to the tools you already run, the software you build, through the API.

Get StartedContact SalesRead the docs