Rename batchTrack endpoint to balances.batch_track

apps/docs/api-reference-generator/core/batchTrack.mdx

@@ -1,6 +1,6 @@
 ---
 title: "Batch Track Usage"
-openapi: "openapi POST /v1/balances.batchTrack"
+openapi: "openapi POST /v1/balances.batch_track"
 ---
 
 import { DynamicParamField } from "/snippets/dynamic-param-field.jsx";

apps/docs/mintlify/api-reference/core/batchTrack.mdx

@@ -1,6 +1,6 @@
 ---
 title: "Batch Track Usage"
-openapi: "openapi POST /v1/balances.batchTrack"
+openapi: "openapi POST /v1/balances.batch_track"
 ---
 
 import { DynamicParamField } from "/snippets/dynamic-param-field.jsx";

apps/docs/mintlify/api/openapi.yml

@@ -16399,7 +16399,7 @@ paths:
                 feature_id="messages",
                 value=1,
             )
-  /v1/balances.batchTrack:
+  /v1/balances.batch_track:
     post:
       operationId: batchTrack
       description: >-
@@ -16492,7 +16492,7 @@ paths:
             ]);
         - lang: curl
           source: |-
-            curl -X POST https://api.useautumn.com/v1/balances.batchTrack \
+            curl -X POST https://api.useautumn.com/v1/balances.batch_track \
               -H 'Authorization: Bearer <token>' \
               -H 'Content-Type: application/json' \
               -d '[

apps/docs/mintlify/changelog/changelog.mdx

@@ -9,7 +9,7 @@ description: "Some new things we've shipped at Autumn HQ"
 
   Two new ways to record usage when you don't need an immediate balance read:
 
-  - [`POST /v1/balances.batchTrack`](/api-reference/core/batchTrack) — enqueue up to **1000 usage events** in one request. Returns 202 immediately; balances are deducted by background workers.
+  - [`POST /v1/balances.batch_track`](/api-reference/core/batchTrack) — enqueue up to **1000 usage events** in one request. Returns 202 immediately; balances are deducted by background workers.
   - **`async: true`** on the existing [`POST /v1/balances.track`](/api-reference/core/track) — same fire-and-forget shape for single-event callers that want the speed without switching to the batch endpoint.
 
   Both paths are intended for high-volume metering (event logging, per-action usage counters) where you'd previously be limited by the synchronous deduction's HTTP round-trip. Partial enqueue failures are logged server-side and do NOT surface as errors to the client — see the [batch track reference](/api-reference/core/batchTrack#partial-failure-semantics) for the trade-off.

packages/openapi/v2.3/contracts/balancesContract.ts

@@ -160,7 +160,7 @@ export const balancesTrackContract = oc
 export const balancesBatchTrackContract = oc
 	.route({
 		method: "POST",
-		path: "/v1/balances.batchTrack",
+		path: "/v1/balances.batch_track",
 		operationId: "batchTrack",
 		description: "Track multiple usage events asynchronously.",
 		spec: (spec) =>

server/src/internal/balances/balancesRouter.ts

@@ -34,6 +34,6 @@ balancesRpcRouter.post("/balances.update", ...handleUpdateBalance);
 balancesRpcRouter.post("/balances.delete", ...handleDeleteBalance);
 
 balancesRpcRouter.post("/balances.track", ...handleTrack);
-balancesRpcRouter.post("/balances.batchTrack", ...handleBatchTrack);
+balancesRpcRouter.post("/balances.batch_track", ...handleBatchTrack);
 balancesRpcRouter.post("/balances.check", ...handleCheck);
 balancesRpcRouter.post("/balances.finalize", ...handleFinalizeLock);

server/src/internal/misc/rateLimiter/rateLimitConfigs.ts

@@ -88,7 +88,7 @@ const RATE_LIMIT_ROUTE_GROUPS: RateLimitRouteGroup[] = [
 	},
 	{
 		type: RateLimitType.BatchTrack,
-		patterns: [route({ method: "POST", url: "/v1/balances.batchTrack" })],
+		patterns: [route({ method: "POST", url: "/v1/balances.batch_track" })],
 	},
 	{
 		type: RateLimitType.Check,

server/tests/integration/balances/batch-track/batch-track-contract.test.ts

@@ -1,5 +1,5 @@
 /**
- * Retrospective contract coverage for POST /v1/balances.batchTrack.
+ * Retrospective contract coverage for POST /v1/balances.batch_track.
  *
  * This file is the HTTP-layer contract of record for the batch async-track
  * endpoint. It exercises everything the customer can observe by talking to
@@ -8,7 +8,7 @@
  * Contract (full surface, verbatim from the spec):
  *
  *   New endpoint:
- *     - POST /v1/balances.batchTrack
+ *     - POST /v1/balances.batch_track
  *         request body:  BatchTrackParams = TrackParams[] where 1 <= len <= 1000
  *         response:      202, body { success: true }
  *         auth:          requires Scopes.Balances.Write
@@ -139,7 +139,7 @@ const postBatchTrack = async ({
 		headers.Authorization = authorization;
 	}
 
-	const response = await fetch(`${autumn.baseUrl}/balances.batchTrack`, {
+	const response = await fetch(`${autumn.baseUrl}/balances.batch_track`, {
 		method: "POST",
 		headers,
 		body: JSON.stringify(body),

server/tests/unit/balances/track/batch-track-retry-dedup.test.ts

@@ -37,7 +37,7 @@
  *
  * If this test starts failing:
  *   DO NOT "fix" it by undoing the pin. The fix is to add idempotency
- *   keys to the /v1/balances.batchTrack API contract — at which point
+ *   keys to the /v1/balances.batch_track API contract — at which point
  *   this file should be rotated to assert the NEW dedup contract
  *   (client-supplied keys produce stable MessageDeduplicationId across
  *   client-driven retries). Until then, this behavior is intentional.