AI Operations

Operational status endpoints for monitoring your AI usage. Check rate limits, provider health, session token consumption, and cache state.

Endpoints

Method	Path	Description	Auth	Rate Limit
GET	/ai/operations/rate-limit	Current rate limit status	JWT + Entitlement	60/min
GET	/ai/operations/models/health	AI provider health status	JWT + Entitlement	60/min
GET	/ai/operations/sessions/:sessionId/usage	Token usage for a session	JWT + Entitlement	60/min
GET	/ai/operations/cache/stats	Prompt cache statistics	JWT + Entitlement	60/min

---

GET /ai/operations/rate-limit

Get your current AI rate limit consumption. Returns how many requests remain in the current window and when it resets.

Authentication: JWT Bearer token + active AI entitlement required. Rate limit: 60/min

Request

• Headers: Authorization: Bearer <token>

Response

Response Example

{
  "data": {
    "limit": 60,
    "remaining": 42,
    "resetsAt": "2026-03-17T14:01:00.000Z",
    "scope": "account"
  }
}

Response Fields

Field	Type	Description
limit	number	Maximum requests allowed in the window
remaining	number	Requests remaining in the current window
resetsAt	string	ISO 8601 timestamp when the window resets
scope	string	Rate limit scope (currently account)

Code Examples

cURL

curl https://api.chainabit.com/api/v1/ai/operations/rate-limit \
  -H "Authorization: Bearer $TOKEN"

JavaScript

const res = await fetch(`${BASE_URL}/ai/operations/rate-limit`, {
  headers: { Authorization: `Bearer ${TOKEN}` },
});
const { data } = await res.json();

Python

import requests

res = requests.get(
    f"{BASE_URL}/ai/operations/rate-limit",
    headers={"Authorization": f"Bearer {TOKEN}"},
)
data = res.json()["data"]

---

GET /ai/operations/models/health

Get the health status of AI provider endpoints. Uses the circuit breaker pattern to report provider availability.

Authentication: JWT Bearer token + active AI entitlement required. Rate limit: 60/min

Request

• Headers: Authorization: Bearer <token>

Response

Response Example

{
  "data": [
    {
      "provider": "openai",
      "status": "healthy",
      "circuitState": null,
      "lastCheckedAt": null
    },
    {
      "provider": "anthropic",
      "status": "healthy",
      "circuitState": null,
      "lastCheckedAt": null
    },
    {
      "provider": "google",
      "status": "degraded",
      "circuitState": "half-open",
      "lastCheckedAt": "2026-03-17T13:55:00.000Z"
    }
  ]
}

Response Fields

Field	Type	Description
provider	string	Provider name (openai, anthropic, google)
status	string	healthy, degraded, or unavailable
circuitState	string | null	Circuit breaker state (closed, half-open, open)
lastCheckedAt	string | null	ISO 8601 last health check timestamp

Status meanings:

Status	Circuit State	Description
healthy	closed or absent	Provider is operating normally
degraded	half-open	Provider is being tested after a failure
unavailable	open	Provider is blocked due to repeated failures

Code Examples

cURL

curl https://api.chainabit.com/api/v1/ai/operations/models/health \
  -H "Authorization: Bearer $TOKEN"

JavaScript

const res = await fetch(`${BASE_URL}/ai/operations/models/health`, {
  headers: { Authorization: `Bearer ${TOKEN}` },
});
const { data } = await res.json();

Python

import requests

res = requests.get(
    f"{BASE_URL}/ai/operations/models/health",
    headers={"Authorization": f"Bearer {TOKEN}"},
)
data = res.json()["data"]

---

GET /ai/operations/sessions/:sessionId/usage

Get token usage summary for all AI runs within a session.

Authentication: JWT Bearer token + active AI entitlement required. Rate limit: 60/min

Request

• Headers: Authorization: Bearer <token>
• Path params: sessionId

Response

Response Example

{
  "data": {
    "sessionId": "cm5sess01",
    "totalInputTokens": 4250,
    "totalOutputTokens": 1823,
    "totalTokens": 6073,
    "estimatedCostUsd": 0.018,
    "runCount": 5
  }
}

Response Fields

Field	Type	Description
sessionId	string	The queried session ID
totalInputTokens	number	Total input tokens consumed across all runs
totalOutputTokens	number	Total output tokens consumed across all runs
totalTokens	number	Combined input + output tokens
estimatedCostUsd	number | null	Estimated cost in USD (null if not calculable)
runCount	number	Number of AI runs in the session

Code Examples

Use the id from Create a Session response as $SESSION_ID.

cURL

curl https://api.chainabit.com/api/v1/ai/operations/sessions/$SESSION_ID/usage \
  -H "Authorization: Bearer $TOKEN"

JavaScript

const sessionId = process.env.SESSION_ID; // id of the AI session to inspect

const res = await fetch(`${BASE_URL}/ai/operations/sessions/${sessionId}/usage`, {
  headers: { Authorization: `Bearer ${TOKEN}` },
});
const { data } = await res.json();

Python

import os
import requests

session_id = os.environ["SESSION_ID"]  # id of the AI session to inspect

res = requests.get(
    f"{BASE_URL}/ai/operations/sessions/{session_id}/usage",
    headers={"Authorization": f"Bearer {TOKEN}"},
)
data = res.json()["data"]

---

GET /ai/operations/cache/stats

Get prompt cache statistics for your account. Shows which prompt compilation layers are currently cached.

Authentication: JWT Bearer token + active AI entitlement required. Rate limit: 60/min

Request

• Headers: Authorization: Bearer <token>

Response

Response Example

{
  "data": {
    "cachedLayers": 3,
    "layers": ["preferences", "persona", "session"]
  }
}

Response Fields

Field	Type	Description
cachedLayers	number	Number of cached prompt layers
layers	string[]	Names of cached layers

Possible layers:

Layer	Description
preferences	Your language and tone preferences
persona	Active persona profile
twin	Digital twin context
session	Session-scoped conversation state
memory	Long-term memory context

Code Examples

cURL

curl https://api.chainabit.com/api/v1/ai/operations/cache/stats \
  -H "Authorization: Bearer $TOKEN"

JavaScript

const res = await fetch(`${BASE_URL}/ai/operations/cache/stats`, {
  headers: { Authorization: `Bearer ${TOKEN}` },
});
const { data } = await res.json();

Python

import requests

res = requests.get(
    f"{BASE_URL}/ai/operations/cache/stats",
    headers={"Authorization": f"Bearer {TOKEN}"},
)
data = res.json()["data"]