Stitch · Rate Limits

Stitch Money Rate Limits

Stitch exposes a single GraphQL endpoint at https://api.stitch.money/graphql. Specific per-client request rate limits and GraphQL query complexity / depth limits are not publicly published; they are governed per client and tier and confirmed during onboarding. Throttled requests are expected to surface as HTTP 429. Idempotency on write operations (disbursements, refunds) is enforced via a required nonce that de-duplicates retried requests rather than via a rate limit. Per-client values are not reconciled in this artifact.

Stitch Money Rate Limits is the machine-readable rate-limit profile for Stitch on the APIs.io network, conforming to the API Commons Rate Limits specification.

It captures 3 rate-limit definitions, measuring requests, complexity, and items.

The profile also includes 3 backoff/retry policies defined and response codes documented for throttled.

Tagged areas include Payments, Open Banking, Pay By Bank, GraphQL, and Africa.

3 Limits Throttle: 429
PaymentsOpen BankingPay By BankGraphQLAfricaSouth AfricaFintechRate LimitingQuotasThrottling

Limits

GraphQL Requests Per Client client
requests
see provider documentation
Per-client request rate against POST /graphql; not publicly published, confirmed at onboarding.
Query Complexity / Depth request
complexity
see provider documentation
GraphQL query depth and complexity may be bounded server-side; request only the fields needed.
Pagination Page Size request
items
cursor-based (first / after)
Relay connections are paginated with first / after; large result sets must be paged.

Policies

Idempotency via Nonce
Disbursement and refund mutations require a unique nonce so retried requests do not create duplicate transactions.
Backoff Strategy
Clients should implement exponential backoff with jitter and honor Retry-After on HTTP 429.
Minimal Field Selection
Query only the precise fields needed to keep responses small and avoid complexity limits.

Sources