ΛXIØM Lens

Governed AI calls. Receipts your product can act on.

AXIØM Lens sits between your application and every model it calls, governing the request, returning a branchable decision, and recording what happened.

Get API key See a governed decision

OpenAI-compatible · Anthropic-compatible · MCP-ready

Product evidence

See the decision, not just the answer.

Every governed call leaves an operational trail: what came in, what AXIØM saw, what it decided, and what the application should do next.

Instrument panel

Decision 8F42

Incoming objective

Decide whether the launch is ready to move into production.

Governed decision

stop_loop

Iteration budget is exhausted without stronger evidence.

Call shapeBusiness decision

Hidden axisMotion without progress

Evidence stateNo new evidence

App actionReturn to human

Decision receipt

Stored

Input came in

Draft a customer-facing claim from incomplete evidence.

AXIØM saw

The request crosses a proof boundary before publication.

AXIØM decided

require_verification

App should do next

Hold the claim, request evidence, and store this receipt.

Governed boundary

Trace visible

Request

received

Pre-lens

contracted

Model

routed

Governor

verified

Receipt

persisted

Governance stays separate from model choice, while every stage remains inspectable by the host application.

Provider routing

Team policy

RouteProviderConstraintStatus

byokCustomer keyInside policyavailable

frontier-managedOpenRouterAllowance enforcedactive

axiom-consensusMulti-modelPolicy selectedavailable

For developers

One governed API. Every model.

Keep the OpenAI or Anthropic client your application already uses. Change the endpoint, preserve the governance envelope, and branch on the returned decision.

OpenAI and Anthropic-compatible HTTP routes

MCP access for IDEs, agents, and tools

Receipts linked to the team decision ledger

Get API key Read documentation

Governed request

curl https://api.axiomlens.com
/v1/chat/completions \
  -H "Authorization: Bearer axm_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "max_tokens": 256,
    "messages": [
      {
        "role": "user",
        "content": "Should this deployment move to production?"
      }
    ]
  }'

Responserequire_verification

{
  "choices": [{
    "message": {
      "role": "assistant",
      "content": "Hold the launch until rollback evidence is attached."
    }
  }],
  "axiom_governance": {
    "action": "require_verification",
    "reason": "production boundary requires rollback proof",
    "passed_all": false,
    "receipt_id": "rcpt_8f42"
  }
}

API formats

OpenAI and Anthropic compatible

Route policies

BYOK, managed, and consensus

Branchable decisions

From allow through stop_loop

Team boundary

API, MCP, usage, and receipts

Commercial access

Choose how provider access is supplied.

The governance boundary stays the same. Your team decides who owns the underlying model account and how usage is constrained.

Self-serve

Bring your own provider keys.

Keep your existing OpenRouter or Anthropic account while AXIØM governs requests and records the resulting decisions.

Team API keys and MCP access

Provider credentials remain under your control

Governance receipts in the decision ledger

Create an account

Managed Frontier

Use provider access supplied by AXIØM.

Run through AXIØM-managed routing under your plan allowance, with quota enforcement and attributed usage visible in Billing.

No separate provider account required

Allowance and spend controls before model calls

The same governed API and MCP boundary

View access options

API keys and MCP access live under the same AXIØM Lens team boundary.

Set up MCP