Start free →

API reference

Gateway calls

Call a model through the governed gateway — and around a call: whether it needs a model, the model a first call can use, resuming a held call, asking for an exception, grounding checks and your own vault’s receipts.

All of the API reference

Route summaries and field names are written in English, here and in the file.

/api/v1/chat

POST /api/v1/chat/completions

The governed gateway, OpenAI-compatible

Authentication
API key
Key scope
inference:write
Headers
X-Azoth-Usecase · X-Azoth-Decision-Ref · X-Azoth-Subject-Zone · X-Azoth-End-User · X-Azoth-Challenge-Passed · X-Azoth-On-Behalf-Of · X-Azoth-Locale · X-Azoth-Alternative · X-Azoth-Refusal-Ref · Idempotency-Key
Body
model (required) · messages (required) · temperature · max_tokens · stream · azoth_use_cache · azoth_task · azoth_cache_policy · azoth_agent_id · azoth_workflow_id · azoth_compute_os · azoth_sources

POST /api/v1/chat

The governed gateway with an explicit provider, fallbacks and streaming

Authentication
API key
Key scope
inference:write
Headers
X-Azoth-Usecase · X-Azoth-Decision-Ref · X-Azoth-Subject-Zone · X-Azoth-End-User · X-Azoth-Challenge-Passed · X-Azoth-On-Behalf-Of · X-Azoth-Locale · X-Azoth-Alternative · X-Azoth-Refusal-Ref · Idempotency-Key
Body
provider (required) · model (required) · messages (required) · temperature · maxTokens · stream · useCache · task · cachePolicy · agentId · workflowId · fallbackProviders · azureEndpoint · azureDeployment · ollamaBaseUrl · onApprovalRequired · azoth_sources
Answers

/api/v1/infer

POST /api/v1/infer/resume

Run a held call once a reviewer approved it

POST /api/v1/infer/preflight

Score whether a request needs a model call at all

Authentication
API key
Key scope
advisory:read
Module
Served where the deployment runs FEATURE_IAE.
Body
connectorId · request

/api/v1/models

GET /api/v1/models/defaults

Which model a first call can use, per provider

Authentication
API key
Key scope
advisory:read

/api/v1/exceptions

POST /api/v1/exceptions

Ask the rule's owner for a time-limited exception to a refusal

Authentication
API key
Key scope
inference:write
Module
Served where the deployment runs FEATURE_HELPFUL_REFUSALS.
Body
refusal_ref (required) · justification (required) · duration_days · requester_label

GET /api/v1/exceptions/{id}

Where an exception this key asked for stands

Authentication
API key
Key scope
inference:write
Module
Served where the deployment runs FEATURE_HELPFUL_REFUSALS.
Parameters
id (required)

/api/v1/grounding

POST /api/v1/grounding/release

Fetch a high-stakes answer the grounding check held, once approved

Authentication
API key
Key scope
inference:write
Module
Served where the deployment runs FEATURE_GROUNDING_CHECK.
Body
approval_request_id (required)

POST /api/v1/grounding/receipts

Seal the receipt of a grounding check run on your side

Authentication
API key
Key scope
inference:write
Module
Served where the deployment runs FEATURE_GROUNDING_CHECK.
Body
traceId (required) · claims (required) · supported (required) · judgeVersion (required) · claimsVersion (required) · sourcesHash (required) · sourceCount (required) · latencyMs · usecaseId

/api/v1/vault

POST /api/v1/vault/receipts

Seal the receipt of a call kept in your own vault

Authentication
API key
Key scope
inference:write
Module
Served where the deployment runs FEATURE_CLIENT_VAULT.
Headers
X-Azoth-Decision-Ref
Body
traceId (required) · plaintextHash (required) · ciphertextHash (required) · keyId